The Role and Impact of Trade Union Equality Representatives in Britain

The Role and Impact of Trade UnionEquality Representatives in Britainbjir_865 239..262

Nicolas Bacon and Kim Hoque

Abstract

Drawing on data from the first national survey of trade union equality repre-sentatives, this article assesses the role and impact of equality representatives inBritain. While the majority of equality representatives report having had apositive impact on employer equality practice, the analysis also suggests thatequality representative effectiveness might be enhanced via the introduction ofstatutory rights to time off. In addition, it points to the need for unions toinclude equality on the bargaining agenda, attract new representatives to therole, and ensure that equality representatives have the confidence and skills torepresent members effectively within grievance procedures. It also suggests animportant role for the government in encouraging employers to engage inmeaningful dialogue with equality representatives.

1. Introduction

It is generally acknowledged that trade unions have had a significant impactrecently in terms of promoting fair treatment at work and advancing theinterests of disadvantaged workers. Studies have shown, for example, thatequal opportunities (EO) practices are more likely to have been adopted inunionized than non-union workplaces, and outcomes such as pay rates havebeen found to be more equitable in unionized workplaces than elsewhere(Bewley and Fernie 2003; Harcourt et al. 2008; Hoque and Bacon 2010;Metcalf et al. 2001; Noon and Hoque 2001; Walsh 2007). Acknowledging thepotential for unions to influence equality practices positively, the previousLabour government in Britain identified an important role for the unionmovement in contributing towards the effective implementation and deliveryof the recent extension of equality legislation in the areas of flexible working,disability, age, sexual orientation, and religion and belief (Trades Union

Nicolas Bacon is at Nottingham University Business School, Jubilee Campus. Kim Hoque is atBirkbeck, University of London.

bs_bs_banner

British Journal of Industrial Relations doi: 10.1111/j.1467-8543.2011.00865.x50:2 June 2012 0007–1080 pp. 239–262

© Blackwell Publishing Ltd/London School of Economics 2011. Published by Blackwell Publishing Ltd,9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

Congress (TUC) 2009a). It also took some tentative steps to develop furtherthe capacity of unions to promote greater fairness and equality at work(Dickens 2007: 484).

One particular union initiative that the previous government supportedfinancially was the TUC’s equality representative initiative. Following therecommendations of the Women and Work Commission’s (2006) inquiryinto gender inequality in the labour market, the government-backed UnionModernisation Fund financed eight union-led projects to recruit and train500 union equality representatives by mid-2009 (TUC 2009b, 2010). Equalityrepresentatives are a new type of union activist appointed or elected fromexisting employees or representatives in unionized workplaces. Their role isto help promote equality and fairness at work by encouraging employers toimprove equality policies and practices, offering independent advice andguidance on equality issues to employees, and raising the profile of theequality agenda within their unions (BIS 2009a: 1). Although many existinglay representatives have for a long time sought to engage in such activities,the establishment of a specific equality representative role is perhaps indica-tive of the increasing emphasis unions are now placing on equality issues(Colling and Dickens 2001; Gregory and Milner 2009; Heery 2006; Parker2002).

The aim of this article is to offer an evaluation of the TUC’s equalityrepresentative initiative, thereby providing the first representative assessmentof the impact union equality representatives have had in British workplaces.This evaluation is important not least given the role the previous governmentidentified for unions in helping to deliver equality policy and the financialsupport it provided, and also given the resources unions themselves havecommitted to the initiative. It is also important given that equality represen-tatives have not been accorded statutory rights to time off, training or facili-ties to undertake their equality representative duties (ACAS 2008: 4), theprevious government having cited insufficient evidence of equality represen-tative impact and employer support in its rejection of union-sponsoredefforts to have these provisions included in the Equality Act 2010. Hence, itis important to know whether equality representatives are able to have apositive impact in the absence of statutory backing, and also whether thereare forms of support other than statutory backing that might be important indeveloping and supporting the initiative in the future.

Assessing the impact of equality representatives is also important as part ofa broader evaluation of the union modernization agenda, one aim of whichhas been the creation of specialized union positions (such as union learningrepresentatives (ULRs), for example) seeking to achieve new goals by con-centrating on specific issues. Increasing the focus on equality via the intro-duction of specialized equality representatives has the potential to increasethe appeal of unions to diverse and traditionally under-represented groups(Noon and Hoque 2001: 106–7; TUC 2009a: 10–11). The equality represen-tative role may also attract new individuals into union representation,thereby increasing the numbers of females and ethnic minorities (and other

240 British Journal of Industrial Relations

© Blackwell Publishing Ltd/London School of Economics 2011.

traditionally under-represented groups) in lay posts (BIS 2009b: 3). Creatingnew equality representative positions may therefore help bring about internalchange in trade unions and in the process help to develop future generationsof more diverse union representatives (ACAS 2008: 7).

2. Unions, equality and equality representatives

The article’s first main aim is to consider whether equality representativeshave been successful in influencing employer equality practice. Existing theo-ries of union effects point to several reasons why one might expect such aninfluence. According to the collective voice/institutional response model(Freeman and Medoff 1984), positive union effects on equality may resultwhere unions negotiate for enhancements to equality policies and practicesvia equality bargaining (Budd and Mumford 2004; Heery 2006). As Delaneyand Lundy (1996) argue, unions have historically sought to use such voicechannels to encourage employers to adopt formal rules and limit manage-ment discretion. Given this, it is perhaps unsurprising that equality practicessuch as systematic monitoring and reviews of procedures to guard againstdiscrimination are more prevalent in unionized firms than elsewhere (Bewleyand Fernie 2003: 102; Kersley et al. 2006: 248; Walsh 2007: 307). One mightexpect equality representatives to amplify union voice effects of this nature byseeking to ensure that equality issues are established on the union bargainingagenda (Heery 2006: 538) and also by engaging employers in informal dia-logue on equality matters.

In addition, unions might influence employer equality practice via facilita-tion effects. These effects result from union provision of information onequality policies to employees, the support and assistance they offer tomembers wishing to investigate harassment or discrimination complaints,and their representation of members within grievance procedures (Budd andMumford 2004). Should unions be successful in raising employees’ awarenessof their legal rights and of the union support available to them (Dickens 2007:484; Kramer 2008), this is likely to increase the probability that employees willpursue equality-related grievances. This in turn will bring discrimination andharassment problems to the attention of employers, who may subsequentlyrespond by improving equality practice to prevent such problems fromre-emerging in the future. One might anticipate that equality representativeswill augment such facilitation effects given their specific expertise and com-mitment to providing advice and support to employees on equality matters.They will also be ideally positioned to offer advice and assistance to theemployer should they choose to redevelop equality policies in light of emergentproblems.

Hence, there are several reasons why one might expect equality representa-tives to have a significant positive impact on employer equality practice. Theextent to which they are able to do this in the absence of statutory backing,however, remains to be seen. The first aim of the article is to address this issue.

Trade Union Equality Representatives 241


3. Potential influences on the effectiveness of equality representatives — the‘Activity-Support-Characteristics’ (ASC) framework

The second aim of the article is to identify the factors associated with theability of equality representatives to influence employer equality practice.As suggested above, identifying the correlates of equality representativeeffectiveness has potentially important implications in terms of how best todevelop and support the equality representative initiative in the future, espe-cially in the absence of statutory backing. As we argue elsewhere (Bacon andHoque 2011) in our analysis of ULRs, the key determinants of representativeeffectiveness (ULR or otherwise) are likely to be the activities representativesengage in, the support they receive and their characteristics. Hence, wepropose an ASC framework, by which the effectiveness of union representa-tives might be analysed. The analysis conducted here draws on this ASCframework to explore the potential correlates of equality representativeeffectiveness.

Equality Representative Activity

In exploring the activities that are likely to be associated with equality rep-resentative effectiveness, a useful starting point is to consider whether equal-ity representatives are engaging in the sort of activities that are likely toenhance the voice and facilitation effects outlined above. With regard tovoice effects, a key factor might be the extent to which equality representa-tives have been able to establish a meaningful and regular dialogue withmanagers on equality matters (Kirton and Greene 2006: 442). A furtherimportant related activity is whether equality representatives participate onequality committees or forums. Evidence from private firms in the USAsuggests that such committees have a significant impact because they embed‘accountability, authority, and expertise’ and integrate equality and diversityissues across the organization (Kalev et al. 2006: 611). Equality representa-tive participation on such committees might provide an important voicemechanism by which they can influence employer equality practice andmonitor progress on equality initiatives.

With regard to facilitation effects, as outlined above, in instances whereequality representatives have offered information and advice to memberson equality matters, supported members suffering discrimination andharassment, and represented them within grievance procedures, one mightanticipate that this will alert employers to equality problems and thatimprovements to equality practice might emerge as a result (Budd andMumford 2004). The article will test this argument by evaluating whetherequality representatives are more likely to report having influenced employerequality practices positively where they have supported and representedworkers in this manner.

It is unlikely, however, that equality representatives will be able to havean impact via either the voice or the facilitation route unless they are able



to spend sufficient time performing the role. This is considered to be acrucial factor for union representative effectiveness in general (ACAS 2008:12), and for equality representative effectiveness more specifically (BIS2009b: 5; TUC 2009b). Assessing how much time equality representativesare able to spend on the role and the impact of this on their effectiveness istherefore a key issue, especially given that equality representatives were notaccorded statutory rights to time off in the Equality Act 2010. As such, afurther aim of the article is to evaluate the extent to which the number ofhours equality representatives spend performing the role is associated withtheir effectiveness.

Support Given to Equality Representatives

As argued by Colling and Dickens (2001) and Gregory and Milner (2009:125), the likely success of unions in promoting equality practices may dependin part on employer encouragement and support, especially in instanceswhere they are attempting to influence employer-driven diversity initiatives(Kirton and Greene 2006: 445). A number of specific elements of employersupport might be particularly important. For example, equality representa-tives currently do not have statutory rights for access to office facilities(ACAS 2008: 4); hence, they are dependent on the employers’ willingness tomake such facilities available to them. The provision of information onequality matters might also affect the ability of the equality representative toperform the role effectively. According to Heery (2006: 524) and Gregoryand Milner (2009: 128), however, few union representatives report that theirefforts with regard to equality are supported by employers. This in turnsuggests that management support for equality representatives may also belimited, and this, one might anticipate, will restrict their ability to play therole effectively.

Also, in terms of support, equality representative effectiveness might beaffected by the extent to which they operate in a supportive bargaining contextwithin which negotiation or consultation over equality issues occurs (Collingand Dickens 2001; Heery 2006). An analysis of the Workplace EmploymentRelations Survey (WERS) 2004 suggests that the positive union effect onequality practices largely only holds where practices are determined via eitherconsultation or negotiation with the union (Hoque and Bacon 2010; Walsh2007: 306). Hence, it is not union recognition per se that is important butactive union involvement in EO decision making. Extending this argument,the impact of equality representatives might be expected to be greater whereequality practices are determined collectively, given the scope this providesthem with to influence practices indirectly by working with union negotiatingofficers to formulate a bargaining position on equality, and by encouragingnegotiating officers to prioritize equality matters. That said, the occurrence ofequality bargaining is limited (Gregory and Milner 2009: 124; Heery 2006:539–40). Given this, if it does influence equality representative effectiveness,only a minority of equality representatives are likely to benefit from it.



Also with regard to support from the broader bargaining context, theWomen and Work Commission recommended that employers and unions‘re-negotiate recognition agreements to provide time off and facilities forequality reps’ (TUC 2009b: 4). Where such agreements have been reached,these may be important in underpinning the equality representative role.Hence, a further aim of the article will be to identify whether equality repre-sentatives report greater influence on employer equality practice in work-places that have such an agreement in place.

Equality Representative Characteristics

A number of individual characteristics might influence the ability of equalityrepresentatives to perform the role effectively. The first is whether the equal-ity representative has prior experience in a union representative role. Uniteclaims that over half of its equality representatives are new activists (Bennett2009: 446). Such representatives may lack the experience necessary toperform the role effectively, with Heery (2006: 533), for example, reportingthat equality bargaining tends to be the preserve of more experiencedrepresentatives. Against this, however, new representatives may have beenattracted to the role because they have a specific interest in equality issues andmay therefore play the role with particular vigour.

Second, the period of time equality representatives have spent on the role(irrespective of whether or not they are a new activist) could influence theireffectiveness, given that as equality representatives gain experience, they arelikely to accumulate the skills and knowledge necessary to perform the roleeffectively (Heery 2006: 533). This is an important issue given that the equal-ity representative role is relatively new — many equality representatives maynot have been in place for sufficient time yet to have had a positive impact.

A further characteristic that might prove important is whether equalityrepresentatives hold another union post as well as the equality representativepost (hybrid representatives). A key element of several unions’ efforts torecruit the first wave of equality representatives was to encourage existingrepresentatives to take such a hybrid role (BIS 2009b: 4). Arguably,hybrid representatives will focus less on equality issues than will dedicatedequality representatives, given that the other representative activities they areresponsible for may crowd out equality issues. Against this, however, hybridrepresentatives might be more effective given that they are likely to be betterintegrated into broader union networks and structures, and may be able tomobilize the influence this provides when playing the equality representativerole (Moore and Wright 2010). In addition, given that equality representa-tives currently lack statutory backing, it may not be possible to play the rolemeaningfully unless they also play another union role (shop steward or safetyrepresentative, for example) and use a proportion of the rights to time off thisadditional role accords them in which to conduct their equality rep duties(Bennett 2009: 445; Moore and Wright 2010). Hence, a significant proportionof equality representatives for whom the role is their first experience of union



representative activity might be expected to have subsequently taken anotherrepresentative role in order to secure statutory rights to time off.

A final characteristic that might influence equality representative effective-ness is the representative’s gender and ethnicity. As Heery (2006: 522) argues,union equality activities are likely to reflect the characteristics and pre-ferences of the representatives involved, and female and ethnic minorityparticipation in representative union roles is widely regarded as a majorinfluence on the likelihood that unions will promote equality and address theneeds of female and ethnic minority members (Bewley and Fernie 2003: 98;Dickens 2007: 484; Heery and Kelly 1988: 502). By extension, where equalityrepresentatives are female and/or from an ethnic minority background, theymay well play the role with particular vigour.

To summarize, the second aim of the article is to explore the associationbetween a range of factors relating to the activities equality representativesengage in, the support they receive and their characteristics, and the extent towhich they report having influenced employer equality practice positively. Asdiscussed above, this is an important issue given that equality representativeswere denied statutory backing in the Equality Act 2010. The analysis con-ducted here will point to whether statutory backing (in particular with regardto rights to time off to perform the role) would have had a positive effect onequality representative effectiveness and also whether there are other non-statutory forms of support that might be important in developing the equal-ity representative role in the future.

4. Data and method of analysis

The data are drawn from a survey of equality representatives conducted onbehalf of the TUC. The survey was distributed to all equality representativeson the TUC’s database of representatives who had undertaken the TUC’sequality representative training course. In total, 282 surveys were distributedvia this route. In addition, surveys were distributed by Unite and the Publicand Commercial Services Union to their own equality representatives. Thesurvey was distributed in September 2009 with reminders being sent duringOctober 2009. In total, 272 responses were received (a response rate of 46 percent), of which 209 are used in this article, once respondents who stated thatthey are not an equality representative are omitted from the sample and onceobservations with missing data are excluded. The response rate to the surveyis in itself notable. Similar large-scale surveys (of the population of ULRs, forexample) have often yielded response rates in the region of 15 per cent (see,e.g., Bacon and Hoque 2011; Wood and Moore 2005: 8), and this has beeninterpreted by McIlroy (2008) as indicative of high levels of representativeinactivity. Following this line of argument, the response rate of 46 per cent tothe equality representative survey in itself suggests that a large proportion ofequality representatives are playing the role with considerable vigour.

The article’s first aim (to estimate the influence of equality representativeson employer equality practice) is addressed using self-report measures of the



impact that respondents state their equality representative activities have hadon employer practices with regard to gender, race, disability, age, sexualorientation, and religion and belief (on a scale of 1–4 where 1 = none and4 = a lot).

In addressing the article’s second aim, the self-report measures of equalityrepresentative impact described above are used as the dependent variableswithin a multivariate analysis that seeks to identify which of the ASC factorsare associated with equality representative effectiveness. The independentvariables used with regard to the ASC factors are listed in Table A1. Giventhe categorical nature of the dependent variables, ordered probit maximumlikelihood is used. This enables a range of observable characteristics thatmight influence the associations between the ASC factors and the impact ofequality representatives on employer equality practice to be controlled for.The controls used in the equation, which relate to both individual and work-place characteristics, are also listed in Table A1. The coefficients for thesecontrol variables will in themselves be of note given that they will demon-strate whether the impact equality representatives state they have had onemployer equality practice varies by public/private sector, industry sector ororganization size.

Some of the means reported in Table A1 are interesting in their own right,especially given how little is currently known about equality representatives.First, it is notable that only 22 per cent of equality representatives spend5 hours or more a week on the role. This in itself suggests that statutory rightsto time off, denied to equality representatives in the Equality Act 2010, wouldhave played an important role in increasing the proportion of equality rep-resentatives able to spend an appreciable amount of time performing theirduties.

Second, it is notable with regard to employer support for equality repre-sentatives that only just over 3 in 10 equality representatives explicitly believethat managers at their workplace do not value their equality representativeactivities. This figure remains the same for both the public and privatesectors. Also, half of equality representatives feel that managers provide themwith adequate information, while the majority of equality representatives feelthat employers provide them with access to sufficient office space and com-munication equipment. One of the previous government’s justifications fornot providing equality representatives with statutory backing was that therewas insufficient evidence of employer support for the initiative. However, theevidence presented here points to considerable levels of employer support.Hence, had statutory rights for equality representatives been included in theEquality Act, employer objections might not have been as great as it wasassumed they would be.

Also of note is that 85 per cent of equality representatives are ‘hybrid’representatives who currently hold another union post as well as an equalityrepresentative post. As argued above, equality representatives may be unableto play the role meaningfully unless they take another union post (shopsteward or safety representative, for example) and use a proportion of the



facility time attached to this post in which to perform the equality represen-tative role (Bennett 2009: 445). Reflective of this, only 7 per cent of dedicatedequality representatives spend 5 or more hours per week on the equalityrepresentative role in comparison with 24 per cent of hybrid representatives.Also, of the few equality representatives for whom the role is their firstexperience of union representation, 43 per cent have now taken anotherunion role in addition to the equality representative role, possibly as theyhave had to do so in order to receive any statutory rights. One might argue,however, that this could stifle new equality representative recruitment, withmany potential equality representatives being deterred from the role giventhat it would inevitably result in them having to take on other additionalunion responsibilities. Therefore, according statutory time off to equalityrepresentatives would not only allow existing equality representatives tospend more time on the role, but it might also facilitate the recruitment ofequality representatives in greater numbers. It might also help to increase thediversity of the union representative population — perhaps unsurprisingly,given that most equality representatives are drawn from the existing unionrepresentative population (currently less than one-fifth of equality represen-tatives (17 per cent) are new activists); the majority (55 per cent) of equalityrepresentatives are male. However, the potential for the role to increasediversity is demonstrated by the fact that of the few equality representativesfor whom the role is their first experience of union representation, 60 per centare female and 26 per cent are from an ethnic minority background.

The characteristics of the workplaces in which equality representativesare found are also noteworthy. One might expect the large majority ofequality representatives to be located in the public sector, given the relativestrength of public-sector unions and the statutory duty on public bodies topromote equality, and also because this is where existing equality bargain-ing is concentrated (Heery 2006: 523; Kirton and Greene 2006: 432).However, a significant proportion of equality representatives (34 per cent)are located within the private sector. They would also appear to be locatedpredominantly in large workplaces and large organizations, suggesting thatalthough there is a paucity of equality representatives in small and medium-sized enterprises (SMEs), their ‘reach’ may extend across large numbers ofemployees.

5. Results

The Impact of Equality Representatives on Employer Equality Practices

The first main aim of the article is to consider the impact of equality repre-sentatives on employer equality practice with regard to gender, race, disabil-ity, age, sexual orientation, and religion and belief. The results, reported inTable 1, demonstrate that more than three-fifths of equality representativesreport having impacted positively on employer disability practices, overhalf report having impacted positively on employer gender, race and age



practices, and just under half report having impacted positively on sexualorientation and religion and belief practices. While a significant proportionof the equality representatives who report having had a positive impact statethat they have had only ‘a little’ rather than ‘some’ or ‘a lot’ of impact, theresults nevertheless suggest that the majority of equality representatives havehad at least some positive effect. Indeed, only 25 per cent of respondentsreport no impact whatsoever on employer practices with regard to any of thesix equality strands asked about. In addition, only 8 per cent of equalityrepresentatives report having impacted positively on only one area ofemployer equality practice, suggesting that very few equality representativesare ‘single-interest’ representatives seeking to address issues concerning asingle equality strand.

The Factors Associated with Equality Representative Effectiveness

The second main aim of the article is to identify the ASC factors associatedwith reports of equality representative effectiveness. The results of themaximum likelihood ordered probit analysis conducted to address this issueare presented in Table 2.

Turning first to the activities equality representatives engage in, Table 2demonstrates that equality representatives who have contact with any level ofmanagement at least once a month are more likely to report having positivelyinfluenced employer equality practices across all six equality strands than arethose who have less frequent contact. This suggests that direct voice providesan important channel by which equality representatives are able to influenceemployer equality practice. Post-estimation marginal effects calculationssuggest that the differences are sizeable, with equality representatives whohave contact with management at least once a month being 31 per cent morelikely to report having influenced employer gender equality practices posi-tively than are equality representatives whose contact is less frequent.1

TABLE 1The Impact of Equality Representative Activity on Employer Equality Practice (Percentages)

What impact have your equality representativeactivities had on employer equality practices

in the following areas?

A lot Some A little None

Positive impact onGender practices 8 22 25 45Race practices 8 19 26 47Disability practices 13 24 25 37Age practices 6 25 25 43Sexual orientation practices 9 16 24 52Religion and belief practices 6 16 23 54

N = 209.



TA

BL

E2

Fac

tors

Ass

ocia

ted

wit

hth

eA

bilit

yof

Equ

alit

yR

epre

sent

ativ

esto

Influ

ence

Em

ploy

erE

qual

ity

Pra

ctic

e

Influ

ence

ofeq

ualit

yre

pres

enta

tive

onem

ploy

er:

Gen

der

equa

lity

prac

tice

sR

ace

equa

lity

prac

tice

sD

isab

ility

prac

tice

s

Equ

alit

yre

pres

enta

tive

acti

vity

Con

tact

wit

han

yle

velo

fm

anag

emen

tin

role

aseq

ualit

yre

pres

enta

tive

atle

ast

once

am

onth

0.84

7(0

.242

)***

0.69

9(0

.235

)***

0.95

2(0

.233

)***

Doe

sth

eem

ploy

erha

vean

equa

lity

com

mit

tee

orfo

rum

that

the

equa

lity

repr

esen

tati

vere

gula

rly

atte

nds

(ref

eren

ceca

tego

ry:n

oco

mm

itte

eor

foru

m)

Com

mit

tee

orfo

rum

that

the

equa

lity

repr

esen

tati

vedo

esno

tre

gula

rly

atte

nd-0

.051

(0.2

79)

0.30

6(0

.265

)-0

.020

(0.2

62)

Com

mit

tee

orfo

rum

that

the

equa

lity

repr

esen

tati

vere

gula

rly

atte

nds

0.60

4(0

.247

)**

0.40

6(0

.241

)*0.

722

(0.2

42)*

**R

efer

ence

cate

gory

:has

not

inve

stig

ated

/ass

iste

dem

ploy

ees

wit

hin

vest

igat

ions

ofdi

scri

min

atio

nor

hara

ssm

ent

com

plai

nts

inth

epa

st12

mon

ths

orre

pres

ente

dem

ploy

ees

Inve

stig

ated

/ass

iste

dem

ploy

ees

-0.3

34(0

.285

)0.

173

(0.2

78)

-0.0

04(0

.274

)R

epre

sent

edem

ploy

ees

suff

erin

gdi

scri

min

atio

nor

hara

ssm

ent

0.20

6(0

.256

)0.

282

(0.2

57)

0.86

0(0

.251

)***

Fiv

eor

mor

eho

urs

per

wee

ksp

ent

oneq

ualit

yre

pres

enta

tive

acti

viti

es0.

483

(0.2

46)*

*0.

368

(0.2

45)

0.49

3(0

.241

)**

Sup

port

for

equa

lity

repr

esen

tati

ves

Do

you

agre

eor

disa

gree

that

man

ager

sat

this

wor

kpla

ceva

lue

your

equa

lity

repr

esen

tati

veac

tivi

ties

(ref

eren

ceca

tego

ry:d

isag

ree)

?N

eith

erag

ree

nor

disa

gree

0.35

9(0

.248

)0.

339

(0.2

42)

0.37

5(0

.236

)A

gree

0.15

3(0

.272

)0.

195

(0.2

69)

0.32

0(0

.267

)E

mpl

oyer

prov

ides

Suffi

cien

tof

fice

spac

e-0

.322

(0.2

67)

-0.1

23(0

.261

)-0

.258

(0.2

61)

Suffi

cien

tco

mm

unic

atio

neq

uipm

ent

-0.4

15(0

.339

)-0

.021

(0.3

18)

0.04

2(0

.317

)A

dequ

ate

info

rmat

ion

toco

nduc

tth

ero

le-0

.051

(0.2

48)

0.20

0(0

.244

)0.

176

(0.2

41)

Arr

ange

men

tsfo

req

ualit

yre

pres

enta

tive

sse

tou

tin

form

alag

reem

ent

0.00

0(0

.226

)0.

130

(0.2

19)

0.27

3(0

.221

)W

hen

deci

ding

equa

lity

polic

ies

and

prac

tice

s,m

anag

ers

norm

ally

(ref

eren

ceca

tego

ry:

dono

tin

volv

eun

ion

repr

esen

tati

ves

atal

l)In

form

0.81

6(0

.339

)**

0.35

1(0

.326

)0.

115

(0.3

13)

Con

sult

1.03

6(0

.333

)***

0.46

8(0

.303

)-0

.089

(0.3

03)

Neg

otia

te1.

213

(0.3

28)*

**0.

808

(0.3

08)*

**0.

292

(0.3

03)



Tab

le2

(con

td)

Influ

ence

ofeq

ualit

yre

pres

enta

tive

onem

ploy

er:

Gen

der

equa

lity

prac

tice

sR

ace

equa

lity

prac

tice

sD

isab

ility

prac

tice

s

Equ

alit

yre

pres

enta

tive

char

acte

rist

ics

Uni

onpo

stpr

evio

usly

held

-0.7

39(0

.305

)**

-0.5

25(0

.293

)*-0

.362

(0.2

93)

Oth

erun

ion

post

curr

entl

yhe

ld(‘

hybr

id’r

epre

sent

ativ

es)

0.72

4(0

.331

)**

0.31

0(0

.328

)0.

040

(0.3

29)

Tim

esp

ent

aseq

ualit

yre

pres

enta

tive

(ref

eren

ceca

tego

ry:l

ess

than

1ye

ar)

1to

less

then

2ye

ars

-0.2

93(0

.253

)-0

.299

(0.2

54)

-0.1

98(0

.245

)2

tole

ssth

an3

year

s0.

012

(0.2

85)

-0.0

21(0

.279

)-0

.206

(0.2

75)

3or

mor

eye

ars

0.48

9(0

.268

)*0.

070

(0.2

62)

-0.3

83(0

.262

)F

emal

e0.

168

(0.2

04)

-0.0

06(0

.197

)-0

.200

(0.1

96)

Eth

nic

min

orit

y-0

.523

(0.2

68)*

0.17

4(0

.244

)-0

.641

(0.2

55)*

*C

ontr

ols

Sect

or(r

efer

ence

cate

gory

:pub

licse

ctor

)P

riva

tese

ctor

-0.4

11(0

.329

)-0

.020

(0.3

18)

0.09

3(0

.312

)N

otfo

rpr

ofit/

volu

ntar

yse

ctor

-0.8

20(0

.640

)-0

.470

(0.6

24)

-1.2

91(0

.623

)**

Org

aniz

atio

nsi

ze(r

efer

ence

cate

gory

:0–9

9em

ploy

ees)

100–

249

empl

oyee

s0.

391

(0.7

79)

-0.8

39(0

.889

)0.

542

(0.7

57)

250–

999

empl

oyee

s-0

.243

(0.6

63)

-0.6

50(0

.664

)-0

.080

(0.6

48)

1,00

0–4,

999

empl

oyee

s-0

.411

(0.6

00)

-0.6

69(0

.588

)-0

.208

(0.5

81)

5,00

0–9,

999

empl

oyee

s-0

.644

(0.6

29)

-1.1

22(0

.626

)*-0

.397

(0.6

06)

10,0

00em

ploy

ees

orm

ore

-0.3

49(0

.597

)-0

.525

(0.5

91)

-0.0

79(0

.581

)St

anda

rdIn

dust

rial

Cla

ssifi

cati

onm

ajor

grou

p(r

efer

ence

cate

gory

:pub

licad

min

istr

atio

nan

dde

fenc

e)E

duca

tion

0.50

3(0

.366

)0.

077

(0.3

61)

0.03

4(0

.356

)H

ealt

han

dso

cial

wor

k-0

.291

(0.3

95)

-0.0

15(0

.386

)-0

.421

(0.3

84)

Tra

nspo

rt,s

tora

gean

dco

mm

unic

atio

ns0.

041

(0.3

33)

0.39

1(0

.319

)-0

.017

(0.3

16)

Fin

anci

alin

term

edia

tion

0.31

2(0

.423

)0.

067

(0.4

26)

-0.2

46(0

.420

)M

anuf

actu

ring

0.90

3(0

.472

)*0.

169

(0.4

72)

0.24

5(0

.450

)O

ther

com

mun

ity,

soci

alan

dpe

rson

alse

rvic

es0.

257

(0.3

37)

0.07

1(0

.336

)0.

012

(0.3

32)

Oth

erin

dust

rygr

oup

0.64

8(0

.534

)0.

279

(0.5

28)

-0.0

36(0

.501

)P

seud

oR

20.

226

0.18

00.

226

N20

920

920

9



Tab

le2

(con

td)

Influ

ence

ofeq

ualit

yre

pres

enta

tive

onem

ploy

er:

Age

equa

lity

prac

tice

sS

exua

lori

enta

tion

equa

lity

prac

tice

sR

elig

ion

orbe

lief

equa

lity

prac

tice

s

Equ

alit

yre

pres

enta

tive

acti

vity

Con

tact

wit

han

yle

velo

fm

anag

emen

tin

role

aseq

ualit

yre

pres

enta

tive

atle

ast

once

am

onth

0.57

2(0

.237

)**

0.60

6(0

.253

)**

0.73

1(0

.251

)***

Doe

sth

eem

ploy

erha

vean

equa

lity

com

mit

tee

orfo

rum

that

the

equa

lity

repr

esen

tati

vere

gula

rly

atte

nds

(ref

eren

ceca

tego

ry:n

oco

mm

itte

eor

foru

m)?

Com

mit

tee

orfo

rum

that

the

equa

lity

repr

esen

tati

vedo

esno

tre

gula

rly

atte

nd0.

367

(0.2

67)

0.16

2(0

.288

)0.

222

(0.2

80)

Com

mit

tee

orfo

rum

that

the

equa

lity

repr

esen

tati

vere

gula

rly

atte

nds

0.58

4(0

.243

)**

0.54

7(0

.257

)**

0.65

3(0

.255

)***

Ref

eren

ceca

tego

ry:h

asno

tin

vest

igat

ed/a

ssis

ted

empl

oyee

sw

ith

inve

stig

atio

nsof

disc

rim

inat

ion

orha

rass

men

tco

mpl

aint

sin

past

12m

onth

sor

repr

esen

ted

empl

oyee

sIn

vest

igat

ed/a

ssis

ted

empl

oyee

s0.

288

(0.2

78)

-0.4

02(0

.306

)-0

.011

(0.3

02)

Rep

rese

nted

empl

oyee

ssu

ffer

ing

disc

rim

inat

ion

orha

rass

men

t0.

678

(0.2

58)*

**0.

079

(0.2

72)

0.32

4(0

.273

)F

ive

orm

ore

hour

spe

rw

eek

spen

ton

equa

lity

repr

esen

tati

veac

tivi

ties

0.81

0(0

.243

)***

0.63

5(0

.258

)**

0.46

8(0

.253

)*S

uppo

rtfo

req

ualit

yre

pres

enta

tive

sD

oyo

uag

ree

ordi

sagr

eeth

atm

anag

ers

atth

isw

orkp

lace

valu

eyo

ureq

ualit

yre

pres

enta

tive

acti

viti

es(r

efer

ence

cate

gory

:dis

agre

e)?

Nei

ther

agre

eno

rdi

sagr

ee0.

182

(0.2

39)

0.50

5(0

.268

)*0.

013

(0.2

57)

Agr

ee0.

180

(0.2

68)

0.69

5(0

.288

)**

0.00

4(0

.278

)E

mpl

oyer

prov

ides

Suffi

cien

tof

fice

spac

e-0

.187

(0.2

64)

-0.2

10(0

.282

)-0

.624

(0.2

76)*

*Su

ffici

ent

com

mun

icat

ion

equi

pmen

t-0

.152

(0.3

24)

-0.4

73(0

.355

)-0

.043

(0.3

28)

Ade

quat

ein

form

atio

nto

cond

uct

the

role

0.34

6(0

.245

)0.

185

(0.2

61)

0.45

8(0

.264

)*A

rran

gem

ents

for

equa

lity

repr

esen

tati

ves

set

out

info

rmal

agre

emen

t0.

075

(0.2

21)

-0.2

32(0

.235

)0.

215

(0.2

30)

Whe

nde

cidi

ngeq

ualit

ypo

licie

san

dpr

acti

ces

man

ager

sno

rmal

ly(r

efer

ence

cate

gory

:do

not

invo

lve

unio

nre

pres

enta

tive

sat

all)

Info

rm0.

512

(0.3

27)

0.60

0(0

.353

)*0.

212

(0.3

44)

Con

sult

0.42

8(0

.312

)0.

473

(0.3

35)

0.26

4(0

.322

)N

egot

iate

0.75

1(0

.311

)**

0.99

9(0

.336

)***

0.87

0(0

.322

)***

Equ

alit

yre

pres

enta

tive

char

acte

rist

ics

Uni

onpo

stpr

evio

usly

held

-0.7

40(0

.298

)**

-0.6

10(0

.326

)*-0

.472

(0.3

14)

Oth

erun

ion

post

curr

entl

yhe

ld(‘

hybr

id’r

epre

sent

ativ

es)

0.45

5(0

.334

)1.

098

(0.3

89)*

**1.

044

(0.3

75)*

**



Tab

le2

(con

td)

Influ

ence

ofeq

ualit

yre

pres

enta

tive

onem

ploy

er:

Age

equa

lity

prac

tice

sS

exua

lori

enta

tion

equa

lity

prac

tice

sR

elig

ion

orbe

lief

equa

lity

prac

tice

s

Tim

esp

ent

aseq

ualit

yre

pres

enta

tive

(ref

eren

ceca

tego

ry:l

ess

than

1ye

ar)

1to

less

then

2ye

ars

-0.4

50(0

.248

)*-0

.088

(0.2

77)

-0.5

54(0

.278

)**

2to

less

than

3ye

ars

-0.1

40(0

.276

)0.

375

(0.3

02)

0.12

5(0

.298

)3

orm

ore

year

s-0

.433

(0.2

66)

0.40

3(0

.285

)-0

.234

(0.2

78)

Fem

ale

-0.1

31(0

.201

)-0

.381

(0.2

17)*

-0.0

80(0

.217

)E

thni

cm

inor

ity

-0.3

54(0

.259

)-0

.447

(0.2

78)

0.01

9(0

.257

)C

ontr

ols

Sect

or(r

efer

ence

cate

gory

:pub

licse

ctor

)P

riva

tese

ctor

0.32

7(0

.322

)-0

.209

(0.3

35)

0.02

4(0

.335

)N

otfo

rpr

ofit/

volu

ntar

yse

ctor

-0.5

01(0

.612

)-0

.106

(0.7

20)

-0.3

83(0

.725

)O

rgan

izat

ion

size

(ref

eren

ceca

tego

ry:0

–99

empl

oyee

s)10

0–24

9em

ploy

ees

-0.8

62(0

.788

)-1

.533

(0.8

21)*

0.03

7(0

.822

)25

0–99

9em

ploy

ees

-1.2

03(0

.641

)*-0

.870

(0.6

76)

-0.6

81(0

.688

)1,

000–

4,99

9em

ploy

ees

-0.7

59(0

.563

)-1

.750

(0.6

28)*

**-0

.761

(0.6

26)

5,00

0–9,

999

empl

oyee

s-1

.203

(0.6

41)*

*-1

.856

(0.6

55)*

**-0

.967

(0.6

56)

10,0

00em

ploy

ees

orm

ore

-0.9

49(0

.567

)*-1

.301

(0.6

17)*

*-0

.603

(0.6

24)

Stan

dard

Indu

stri

alC

lass

ifica

tion

maj

orgr

oup

(ref

eren

ceca

tego

ry:p

ublic

adm

inis

trat

ion

and

defe

nce)

Edu

cati

on0.

056

(0.3

58)

-0.5

91(0

.441

)-0

.295

(0.3

93)

Hea

lth

and

soci

alw

ork

-0.1

81(0

.382

)-0

.442

(0.4

18)

-0.3

31(0

.401

)T

rans

port

,sto

rage

and

com

mun

icat

ions

-0.0

17(0

.323

)0.

036

(0.3

39)

-0.0

04(0

.342

)F

inan

cial

inte

rmed

iati

on-0

.051

(0.4

29)

-0.2

50(0

.453

)-0

.351

(0.4

61)

Man

ufac

turi

ng0.

278

(0.4

56)

0.43

1(0

.494

)-0

.226

(0.4

99)

Oth

erco

mm

unit

y,so

cial

and

pers

onal

serv

ices

-0.2

36(0

.334

)-0

.576

(0.3

72)

-0.1

51(0

.366

)O

ther

indu

stry

grou

p-0

.123

(0.5

38)

0.06

6(0

.553

)0.

039

(0.5

70)

Pse

udo

R2

0.20

00.

240

0.22

8N

209

209

209

Not

es:

Coe

ffici

ents

give

n.St

anda

rder

rors

inbr

acke

ts.O

rder

edpr

obit

anal

ysis

.Equ

atio

nsal

soco

ntro

lfor

wor

kpla

cesi

zean

dw

heth

erth

eeq

ualit

yre

pres

enta

tive

repr

esen

tsm

anag

ers/

prof

essi

onal

s,no

n-m

anag

ers

ora

com

bina

tion

ofbo

thgr

oups

.**

*Si

gnifi

cant

at1

per

cent

;**

Sign

ifica

ntat

5pe

rce

nt;*

Sign

ifica

ntat

10pe

rce

nt.



Also in terms of direct voice effects, Table 2 suggests that equality repre-sentatives who attend equality committees or forums are more likely thanequality representatives in workplaces without a forum to report havingpositively influenced employer equality practices across all six equalitystrands (although the association is weak where race equality practices areconcerned). The differences are substantial, with the post-estimation mar-ginal effects analysis showing that equality representatives who regularlyattend equality committees or forums are 22 per cent more likely to reporthaving influenced employer gender equality practices positively than areequality representatives in workplaces that do not have an equality commit-tee or forum.

With regard to facilitation effects, there is some support for the argumentthat where equality representatives have helped employees suffering discrimi-nation and harassment and have represented them within grievance proce-dures, this will alert employers to equality problems and improvements toequality practices will emerge as a result. The results suggest that it is repre-sentation that is particularly important, with equality representatives beingmore likely to report having influenced disability and age practices positivelywhere they have represented employees suffering discrimination or harass-ment. In terms of the size of the differences, the post-estimation marginaleffects analysis shows that equality representatives who have representedemployees suffering discrimination or harassment are 30 per cent more likelyto state that they have influenced employer disability practices positivelythan are equality representatives who have neither represented nor assistedemployees with investigations of discrimination or harassment complaints.

Also concerning the activities equality representatives engage in, the resultssuggest that representatives spending 5 or more hours per week performingthe role are more likely to report having influenced employer practices posi-tively with regard to five of the six equality strands than are equality repre-sentatives who spend less time on the role (although the association is weakwhere religion or belief practices are concerned). The marginal effects analy-sis shows that equality representatives who spend 5 or more hours per weekon the role are 18 per cent more likely to state that they have influencedemployer gender equality practices positively than are equality representa-tives who spend less than 5 hours per week on the role.

Turning to the support equality representatives receive, Table 2 points tothe importance of a supportive bargaining environment. Equality represen-tatives in workplaces where negotiation over equality policies and practicesoccurs are more likely to report having influenced employer equality prac-tices positively across five of the six equality strands (the exception beingdisability practices) than are equality representatives in workplaces whereunion representatives are not involved in such decisions. The differences areconsiderable, with equality representatives in workplaces where negotiationover equality occurs being 41 per cent more likely to report having influencedemployer gender equality practices positively than are equality representa-tives in workplaces where representatives are not involved in EO decision



making. It is also notable that it is negotiation rather than consultation thatis important — there is little evidence that equality representatives report agreater influence on employer equality practices in instances where consul-tation occurs than in instances where representatives are not involved in EOdecision making.

Turning to equality representative characteristics, Table 2 suggests thatequality representatives performing another union representative role as wellas the equality representative role (hybrid representatives) are more likely toreport having influenced employer gender, sexual orientation and religion orbelief practices positively than are dedicated equality representatives. Themarginal effects analysis shows that hybrid equality representatives are 28per cent more likely to state that they have influenced employer genderpractices than are dedicated equality representatives. There is also someevidence that equality representatives with no prior experience of unionrepresentation are more likely to report a positive impact than are equalityrepresentatives who held a representative post prior to adopting the role. Inparticular, new representatives are more likely to state that they have influ-enced gender and age practices positively, and are slightly more likely (at the10 per cent significance level) to report having influenced race and sexualorientation practices positively. With regard to the magnitude of the effect,equality representatives with no prior representative experience are 26 percent more likely to report having influenced employer gender equality prac-tice positively than are equality representatives who had a prior representa-tive role.

Finally, some of the controls included in Table 2 are noteworthy. First,given the statutory duty to promote race and gender equality within thepublic sector, one might assume that equality representatives would be betterplaced to influence employer equality practices positively in the public sectorthan in the private sector. There is, however, no evidence to support thisassumption. Second, there is no evidence to suggest that equality represen-tative effectiveness varies by industry sector, or consistent evidence that itvaries by organization size. Although there are very few equality representa-tives within the SME sector currently (organizations with fewer than 250employees), they are as likely to report having influenced employer equalitypractices positively as are their counterparts in larger organizations.

6. Discussion

This article had two main aims. The first aim was to consider the influence ofequality representatives on employer equality practice. The second aim wasto identify the ASC factors associated with reports of equality representativeeffectiveness.

With regard to the first aim, the results demonstrate that the majority ofequality representatives report having had a positive effect. Indeed, three-quarters of equality representatives report a positive influence on at least one



aspect of employer equality practice. It is also notable that their reportedimpact does not vary between the public and private sectors, between SMEsand large organizations or between industry sectors. Hence, the majority ofequality representatives appear to believe that they are playing the role withconsiderable success across all of the areas of the British economy in whichthey are found.

With regard to the article’s second main aim, the analysis revealed thatseveral ASC factors are associated with reports of equality representativeeffectiveness. In terms of the activities equality representatives engage in, theresults point to the importance of activities that are likely to enhance directvoice. In particular, supporting Kirton and Greene (2006: 442), equalityrepresentatives are more likely to report a positive influence where they havebeen successful in encouraging managers to engage in regular dialogue. Alsowith regard to voice effects, the results point to the importance of equalityforums or committees. Given this, it is perhaps concerning, as demonstratedby Table A1, that less than half of equality representatives are in workplaceswith equality forums, and of these, 4 in 10 equality representatives do notattend. Given broader evidence that equality forums have a significantimpact on diversity management (Kalev et al. 2006), the work of equalityrepresentatives might be assisted by encouraging the greater adoption of suchforums and by encouraging managers to ensure that equality representativesare invited to them.

In addition, the results provide some support for the argument thatequality representatives will have positive facilitation effects in instanceswhere they support and represent workers taking equality cases (Budd andMumford 2004). However, it would appear that representing employeestaking equality cases is more important than simply assisting employees, theformer being perhaps likely to have a greater direct effect on employers thanthe latter.

Also in terms of the activities equality representatives engage in, as withother union representative activities (ACAS 2008: 12), equality representa-tives who spend longer on the role are more likely to report having had apositive impact. Given this, it is perhaps concerning that only 22 per cent ofequality representatives are able to spend 5 hours a week or more on equalityrepresentative activities. This suggests that statutory rights to time off, deniedto equality representatives in the Equality Act, would have contributed sig-nificantly towards helping them perform the role more effectively.

In terms of the support equality representatives receive, the bargainingcontext would appear to be particularly important (Colling and Dickens 2001;Heery 2006), with equality representatives being more likely to report havinginfluenced employer equality practices in instances where negotiation overequality takes place. As discussed earlier, negotiation over equality providesequality representatives with the scope to influence employer equality prac-tices indirectly, either by working with union negotiating officers to formulatea bargaining position on equality or by encouraging negotiating officers topromote equality matters. The results here suggest that ‘indirect’ voice effects



of this nature provide an important avenue by which equality repre-sentatives might exercise influence. This in turn implies that a statutory rightto bargain over equality would boost equality representative effectivenesssignificantly. Such a right might have particular potency given that only 26per cent of equality representatives report that negotiation over equalityoccurs in their workplace.2 However, in the current political context in whichequality representatives have not even been accorded rights to time off, theintroduction of a statutory right of this nature is highly unlikely. As such,the onus may be on unions to take steps to ensure that equality is included inthe bargaining agenda. The evidence here suggests that should they be suc-cessful in doing so, they will reap significant benefits in terms of assisting theefforts of equality representatives to improve equality practices within Britishworkplaces.

The analysis also revealed several equality representative characteristicsto be associated with reports of equality representative effectiveness. First,hybrid equality representatives are more likely to report having influencedemployer equality practices than are dedicated equality representatives. Asdiscussed above, given the lack of statutory rights to time off, many equalityrepresentatives would appear to have to use a proportion of the facility timeattached to another representative role in order to carry out the equalityrepresentative role (hence the high number of hybrid equality representa-tives) (Bennett 2009: 445). It is worth keeping in mind, however, that theequation in Table 2 includes a dummy for the amount of time equalityrepresentatives spend performing the role, yet a hybrid representative effectremains despite this. This suggests support for Moore and Wright’s (2010)argument that hybrid representatives are likely to play the role more effec-tively as they are better integrated into broader union networks and struc-tures given their wider union representative activity, and this may providethem with access to resources they can mobilize when playing the equalityrepresentative role. This may also have a spillover effect in that hybridrepresentatives, having been sensitized to equality issues as a result of playingthe equality representative role, may be more likely to take equality issuesinto account in performing their other representative duties, thereby helpingto mainstream equality issues within broader union activity.

Second, the analysis suggests that although most equality representatives(83 per cent) are drawn from the existing body of union representatives, thefew representatives for whom the equality representative role is their firstexperience of union representation report greater effectiveness than thosewho previously held a representative position, thus suggesting that new rep-resentatives will be especially dedicated to the role and hence will play it withparticular vigour. The effectiveness of the equality representative initiativemight therefore be further enhanced if new representatives with a particularinterest in equality issues can be encouraged to take up the role. This isperhaps unlikely to happen, however, given the argument outlined earlierconcerning the potential for the lack of statutory rights to stifle the recruit-ment of new equality representatives.



It is notable, however, that certain equality representative characteristicsare not associated with equality representative effectiveness. For example,the length of time equality representatives have spent on the role is notassociated with their reported influence. Instead, equality representativesappear to be able to have a positive influence very soon after adopting therole. Hence, there are no grounds to argue that the impact of equalityrepresentatives will be limited currently given that the initiative is new andit will take time before its effects become apparent. In addition, contrary tothe arguments made by Bewley and Fernie (2003: 98), Dickens (2007: 484)and Heery and Kelly (1988: 502), there is no consistent evidence that equal-ity representative effectiveness varies by gender or ethnicity — the evidencehere (albeit based on self-reports) suggests that white males are just aseffective as equality representatives as are their female and ethnic minoritycounterparts. That said, one might anticipate that white male equality rep-resentatives will be more committed to equality issues than white malesplaying other union representative roles.

7. Conclusion

Overall, this article has revealed a number of notable findings. It would appearthat although equality representatives may not be as numerous as other newforms of union representatives, many of them are playing the role withconsiderable vigour. There would appear to be considerable managementsupport for the initiative, and equality representatives report having impactedpositively on employer equality practice. They also report as positive aninfluence in the private sector as in the public sector, and in SMEs as in largerorganizations. This suggests that they may have the potential to improveemployer equality practice within all sectors of the British economy.

However, these findings might also be interpreted as suggesting that thereis no particular urgency to provide equality representatives with statutorysupport in the future, given that they appear to have had a positive impacteven in the absence of such backing. There are reasons, though, to cautionagainst reaching this conclusion too readily. First, while the findingssuggest that equality representatives do indeed report a positive impact,they also suggest that this impact would be enhanced further by statutoryrights. In particular, a statutory right to time off, denied to equality rep-resentatives in the Equality Act 2010, would be of significant benefit notjust in terms of boosting the amount of time equality representatives areable to spend performing their duties (a key determinant of equality rep-resentative effectiveness) but also in terms of encouraging greater numbersof people with a particular interest in equality into the role. Second, onemust keep in mind that equality representatives are currently found in rela-tively few workplaces, and given that it is a new initiative, unions may havebeen particularly keen to establish the role in workplaces that are likelyto be receptive to it. Whether equality representatives would be able to



continue to play the role with equal effectiveness in the absence of statutorysupport were the initiative to be rolled out across the economy more widelyremains to be seen.

In developing and supporting the equality representative initiative,however, it is not just statutory rights that are likely to prove important. Theresults suggest that there is also an important role for unions in terms ofensuring that equality is included and prioritized on bargaining agendas,attempting to attract new representatives with a particular interest in equalitymatters to the role, and ensuring that equality representatives have the con-fidence and skills to represent members within disciplinary and grievanceprocedures. The results also point to an important role for the government inenhancing equality representative voice effects by encouraging employers toengage in meaningful dialogue with equality representatives, possibly via theestablishment of equality forums. If the equality representative initiative canbe supported in this manner, this may prove invaluable in helping to ensureits continued future success.

When drawing these conclusions, however, it is important to note certaincaveats and directions for future research. In particular, one must keep inmind that the measures of equality representative impact used here are basedon self-report data, within which there is the possibility that respondents haveexaggerated reports of their influence. As such, there is scope for furtherresearch exploring the precise nature and extent of the influence of equalityrepresentatives on actual workplace practice, perhaps via in-depth qualitativestudies. One must also keep in mind the potential for common method bias— equality representatives who report engagement in activities likely togenerate voice and facilitation effects or report support from managementand the bargaining environment, for example, may have also respondedpositively to questions concerning their influence. This might be addressed byconducting a matched-data analysis within which data relating to equalityrepresentative impact and data relating to potential influences on their effec-tiveness are drawn from different data sources (for an example of researchtaking such an approach, see Hoque and Bacon’s (2011) study of ULRs). Ifanalyses of this nature are able to replicate the results reported here, this willprovide further grounds to argue for the positive effects equality representa-tives are having on employer equality practice in Britain and will providefurther corroborating evidence for the factors that might influence theireffectiveness.

Final version accepted on 10 March 2011.

Acknowledgements

We would like to thank the TUC for funding the research presented in thisarticle and in particular Theresa Daly and Sarah Veale for the help andadvice they provided.



Notes

1. The post-estimation marginal effects analysis discussed here is not reported inTable 2 but is available on request from the authors.

2. This figure is somewhat higher than in the economy as a whole. WERS 2004 shows,for example, that in workplaces with a recognized trade union, negotiation overequality occurs in 15 per cent of workplaces (Kersley et al. 2006: 194). One possibleexplanation for this discrepancy is that unions may have sought to establishequality representative posts in workplaces that have a bargaining context that islikely to be receptive to them. This is perhaps unsurprising given that this is a newinitiative that the union movement will not wish to see fall flat.

References

ACAS (2008). ‘Employee Representatives: Challenges and Changes in the Work-place’. ACAS Policy Discussion Paper.

Bacon, N. and Hoque, K. (2011). ‘Union representation and training: the impact ofunion learning representatives and the factors influencing their effectiveness’.Human Relations, 64 (3): 387–413.

Bennett, T. (2009). ‘New ways of promoting equality and diversity in the workplace:the role of the union equality representative’. Equal Opportunities International, 28(5): 443–7.

Bewley, H. and Fernie, S. (2003). ‘What do unions do for women?’ In H. Gospel andS. Wood (eds.), Representing Workers. London: Routledge, pp. 92–118.

BIS (2009a). TUC — Training Equality Reps. Department for Business Innovationand Skills, URN 09/1517.

—— (2009b).Unite (T&G) Developing and Supporting Workplace Union EqualityRepresentatives. Department for Business Innovation and Skills, URN 09/532.

Budd, J. W. and Mumford, K. (2004). ‘Trade unions and family-friendly policies inBritain’. Industrial and Labor Relations Review, 57 (2): 204–22.

Colling, T. and Dickens, L. (2001). ‘Gender equality and trade unions: a new basis formobilisation?’ In M. Noon and E. Ogbonna (eds.), Equality, Diversity and Disad-vantage in Employment. Basingstoke: Palgrave, pp. 136–55.

Delaney, J. T. and Lundy, M. C. (1996). ‘Unions, collective bargaining, and thediversity paradox’. In E. Kossek and S. Lobel (eds.), Managing Diversity. Cam-bridge, MA: Blackwell, pp. 245–72.

Dickens, L. (2007). ‘The road is long: thirty years of equality legislation in Britain’.British Journal of Industrial Relations, 45 (3): 463–94.

Freeman, R. and Medoff, J. (1984). What Do Unions Do? New York: Basic Books.Gregory, A. and Milner, S. (2009). ‘Trade unions and work-life balance: changing

times in France and the UK?’ British Journal of Industrial Relations, 47 (1): 122–46.Harcourt, M., Lam, H., Harcourt, S. and Flynn, M. (2008). ‘Discrimination in hiring

against immigrants and ethnic minorities: the effect of unionization’. InternationalJournal of Human Resource Management, 19 (1): 98–115.

Heery, E. (2006). ‘Equality bargaining: where, who, why?’ Gender, Work and Orga-nization, 13: 522–42.

—— and Kelly, J. (1988). ‘Do female representatives make a difference? Womenfull-time officials and trade union work’. Work, Employment and Society, 2 (4):487–505.



Hoque, K. and Bacon, N. (2010). ‘Unions and Workplace Equality in Britain: Evi-dence from the 2004 Workplace Employment Relations Survey’. Paper presented atthe Work Employment and Society Conference, Brighton, September 7–9.

—— and —— (2011). ‘Assessing the impact of union learning representatives ontraining: evidence from a matched sample of ULRs and managers’. Work, Employ-ment and Society, forthcoming.

Kalev, A., Dobbin, F. and Kelly, E. (2006). ‘Best practices or best guesses? Assessingthe efficacy of corporate affirmative action and diversity policies’. American Socio-logical Review, 71 (4): 589–617.

Kersley, B., Alpin, C., Forth, J., Bryson, A., Bewley, H., Dix, G. and Oxenbridge, S.(2006). Inside the Workplace: Findings from the 2004 Workplace Employment Rela-tions Survey. London: Routledge.

Kirton, G. and Greene, A. M. (2006). ‘The discourse of diversity in unionised con-texts: views from trade union equality officers’. Personnel Review, 35 (4): 431–48.

Kramer, A. (2008). ‘Unions as facilitators of employment rights: an analysis ofindividuals’ awareness of parental leave in the National Longitudinal Survey ofYouth’. Industrial Relations, 47 (4): 651–8.

McIlroy, J. (2008). ‘Ten years of New Labour: workplace learning, social partnershipand union revitalization in Britain’. British Journal of Industrial Relations, 46 (2):283–313.

Metcalf, D., Hansen, K. and Charlwood, A. (2001). ‘Unions and the sword of justice:unions and pay systems, pay inequality, pay discrimination and low pay’. NationalInstitute Economic Review, 176 (1): 61–75.

Moore, S. and Wright, T. (2010). PCS Equality Representatives’ Training, Develop-ment & Support Programme. Final Evaluation Report. Working Lives ResearchInstitute, London Metropolitan University.

Noon, M. and Hoque, K. (2001). ‘Ethnic minorities and equal treatment: the impactof gender, equal opportunities policies and trade unions’. National Institute Eco-nomic Review, 176 (1): 105–27.

Parker, J. (2002). ‘Women’s groups in British unions’. British Journal of IndustrialRelations, 40 (1): 23–48.

TUC (2009a). TUC Equality Reps Project Report. London: Trades Union Congress.—— (2009b).TUC Equality Audit 2009. London: Trades Union Congress.—— (2010). Equality Reps Make a Real Difference to UK Workplaces. Press

Release, 21 January.Walsh, J. (2007). ‘Equality and diversity in British workplaces: the 2004 Workplace

Employment Relations Survey’. Industrial Relations Journal, 38 (4): 303–19.Women and Work Commission (2006). Shaping a Fairer Future. Department of

Trade and Industry, URN 06/697.Wood, H. and Moore, S. (2005). An Evaluation of the UK Union Learning Fund —

Its Impact on Unions and Employers, Summary Report. Working Lives ResearchInstitute, London Metropolitan University.



Appendix

TABLE A1Means of Independent and Control Variables

Equality representative activityContact with any level of management in role as equality representative at leastonce a month

0.373

Does the employer have an equality committee or forum which the equalityrepresentative regularly attends?:

No committee or forum 0.531Committee or forum that the equality representative does not regularly attend 0.187Committee or forum that the equality representative regularly attends 0.282

Has not investigated/assisted employees with investigations of discrimination orharassment complaints in past 12 months or represented employees:

0.273

Investigated/assisted employees 0.196Represented employees suffering discrimination or harassment 0.531

Five or more hours per week spent on equality representative activities 0.215Support for equality representatives

Do you agree or disagree that managers at this workplace value your equalityrepresentative activities:

Disagree 0.316Neither agree nor disagree 0.388Agree 0.297

Employer provides:Sufficient office space 0.660Sufficient communication equipment 0.794Adequate information to conduct the role 0.502

Arrangements for equality representatives set out in formal agreement 0.254When deciding equality policies and practices managers normally:

Do not involve union representatives at all 0.220Inform 0.163Consult 0.359Negotiate 0.258

Equality representative characteristicsUnion post previously held 0.833Other union post currently held (‘hybrid’ representatives) 0.852Time spent as equality representative:

Less than one year 0.3011 to less than 2 years 0.2782 to less than 3 years 0.1963 or more years 0.225

Female 0.445Ethnic minority 0.182

Control variablesPublic sector 0.636Private sector 0.335Not for profit/voluntary sector 0.028Organisation size:

0–99 employees 0.034100–249 employees 0.024250–999 employees 0.0671000–4999 employees 0.2445000–9999 employees 0.15810000 employees or more 0.474

Workplace size:�50 employees 0.12051–100 employees 0.072101–250 employees 0.139251–500 employees 0.177



Table A1 (contd)

501–750 employees 0.086751–1000 employees 0.029>1000 employees 0.378

Standard Industrial Classification major group:Public administration and defence 0.291Education 0.086Health and Social Work 0.086Transport, storage and communications 0.177Financial intermediation 0.101Manufacturing 0.110Other community, social and personal services 0.105Other industry group 0.043

Occupational group represented:Managers/professionals 0.115Non-management 0.589Mixed 0.297

Representative age:�29 years old 0.03430–39 years old 0.17240–49 years old 0.38850–59 years old 0.345�60 years old 0.062

N = 209.



Copyright of British Journal of Industrial Relations is the property of Wiley-Blackwell and its content may not

be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.

The Role and Impact of Trade Union Equality Representatives in Britain

Documents