Does Union Membership Benefit Immigrant Workers in Hard Times?
Evidence from Ireland
Tom Turner, Christine Cross and Michelle O’Sullivan
Introduction
Understanding labour force dynamics in the context of increasing immigration levels is an
issue that is of concern to organisations, trade unions and policy makers. This is particularly
true in Europe where despite the current global economic recession countries will
increasingly depend on immigrant labour in the 21st century to fill vacant positions. Driving
this need is a combination of demographic changes, the highly segmented labour market and
differentiated economies (OECD, 2003; Hix and Noury, 2007; Finch et al. 2009). Within
Europe, Ireland is an interesting case in relation to inward migration of labour. Since the late
1990s Ireland experienced rapid economic growth and a corresponding expansion in
employment. These developments changed Ireland from a country of emigration to one of
net immigration. Most of the inward flows can be attributed to the Irish government’s
decision not to restrict entry of the 10 EU accession States in 2004. Non-Irish nationals now
comprise a substantial portion of the population. While they accounted for just 2 per cent of
the employed labour force in 1994, this figure reached 16 per cent by 2006, only declining to
12 per cent in 2010 during the course of the current economic recession (CSO, 2006, 2011).
A growth in immigrant workers raises important issues for trade unions. Immigrant
workers tend to be more likely than national employees to work in precarious employments,
characterised by lower pay, job insecurity, poor job quality and a lack of control of work
tasks (McKay et al., 2011; Cranford et al., 2003; Turner and O’Sullivan, forthcoming). Such
jobs can be difficult to unionise as Immigrant workers can be fearful of voicing concerns over
their pay and conditions particularly where the work is insecure and there is employer
hostility (Dundon et al., 2007). The possibility that immigrants will work for lower pay may
be a concern to trade unions which fear that this will lead to unemployment for their members
and a ‘race to the bottom’ in terms of wages (Dundon et al., 2007; Krings, 2009). These
problems facing both immigrants and national workers can arguably be alleviated by
immigrants joining unions and benefiting from better wages and working conditions.
Immigrant workers present a potential pool of new members for trade unions that could partly
stem the significant declines in membership and density. Union density has been declining
across the European Union, from 27.8% in the EU27 in 2000, to 23.4% in 2008 (European
Commission 2011). The fall in union membership and density has occurred in most countries
but has tended to be smaller in those countries with strong social democratic parties (Schmitt
and Mitukiewicz 2011). While the number of workers joining unions increased in the boom
period of ‘the Celtic Tiger’ in the late 1990s nevertheless there was a steady decline in union
density. Clearly the proportion of workers joining unions failed to keep pace with the
expansion in the numbers employed in the labour market. Much of this decline occurred in
the private sector of the economy. Union density in the private sector declined from over 40
per cent in 1990 to 22 per cent by 2008 (D’Art et al, 2013). In response to membership
decline trade unions particularly in liberal market economies like Ireland have placed a
greater emphasis on organising activities that target the unorganised including immigrants
(Turner et al, 2011). In order to attract immigrant workers trade unions need to provide
convincing evidence that membership delivers real material benefits.
This paper examines whether union membership conveys such material benefits to
immigrants working in the private sector of the economy. Using the 2008 National
Employment Survey, a matched survey of firms and employees, this paper examines the
unionisation rates of immigrant workers and identifies whether trade union membership and
collective bargaining coverage benefits immigrant workers when compared to Irish nationals
(both union and non-union) in terms of wages and benefits. The paper begins with a review
of the role of trade unions in the workplace followed by a discussion of the relationship
between trade unions and migrant workers.
The Benefits of Union Membership
The classic definition of the function or role of trade unions is provided by the Webbs1 as ‘a
continuous association of wage-earners for the purpose of maintaining and improving the
conditions of their working lives’. Workers embraced trade unions as an instrument with
which to exert some influence on wage determination and check the exercise of absolute and
arbitrary employer power. Collective bargaining evens up the asymmetrical power imbalance
inherent in the employment relationship by increasing the market power of workers to
negotiate wage raises while non-union individual workers rely on individual sources of power
such as skill and expertise. Research indicates that the majority of workers join unions to
1
Beatrice and Sidney Webb (1920) A History of Trade Unions.
improve their pay and working conditions. Union membership is attractive to the extent
that it is instrumental in achieving these goals (Crouch, 1982; Waddington and Whitson,
1997). Conversely dissatisfaction with wages and conditions of employment is likely to
increase the propensity of workers to regard unions more favourably (Hartley, 1992: 169170). This instrumental role of trade unions or in Flanders (1970) terms acting in their ‘vested
interest’ can be gauged with regard to the level of members’ wages and conditions relative to
non-union workers. A major reason for being in a union is the extra pay or premium that
accrues to members particularly in Anglo-Saxon countries2. In the US for example the union
wage premium averaged around at least 17 per cent between 1973 and 2002 (Hirsch, 2004;
Hirsch and Schumacher (2004) Blanchflower and Bryson, 2003). Data for the UK indicates a
lower union wage premium that is declining over time from the mid-1990s reaching at best
10 per cent or lower by 2002 (Blanchflower and Bryson, 2004). Estimates for the wage gap in
Canada for 1999 were 14.4 per cent but this falls to 7.7 per cent when the gap is adjusted for
employee and workplace characteristics (Fang and Verma, 2002: 20). In the Irish case Callan
and Reilly (1993) estimated a union membership mark-up of over 20 per cent for a sample of
male non-agricultural workers (1167) in all sectors based on a 1987 survey of income,
distribution and poverty. A comparative study of nine countries that included Ireland reported
a similar wage gap (Freeman, 1994).
A second core function of trade unions is to act as a ‘sword of justice’ to ensure
fairness and due process in the workplace and often in the wider society. In this role unions’
move beyond the notion that the employment relationship is a purely economic transaction
where market based outcomes are viewed as fair, simply because they are produced by
market exchange (Budd, 2005). Unions attempt to operate within a moral arena that calls for
judgements of fairness and justice in market outcomes such as ‘a fair days pay for a fair days
work’. Essentially unions negotiate the space between moral dispositions and norms and the
apparent economic imperatives of a market society (Sayer, 2000). In this regard trade unions
traditionally affect the shape of the pay structure by ensuring lower levels of income
dispersion among union members compared to non-members (Metcalf et al, 2001; Metcalf,
1982). Unions act to reduce levels of income inequality by raising the wages of workers at
2
The union wage premium has received less attention in Continental Europe partly because in these countries
collective bargaining often has almost universal coverage to include non-union workers as well as the unionised
(Bryson, 2007; Visser, 2006). In five countries France, Germany, Italy, the Netherlands and Sweden the union
wage premium is zero (Bryson, 2007: 39). Studies of union wage premiums in European countries tend to be on
the effect of different bargaining levels (such as multi-employer versus single-employer) on pay dispersion
(Dell’Aringa and Pagani, 2007).
the bottom of the income hierarchy and/or lowering the wages paid to the top earners
(through social and political pressures). One of the most significant ways in which this is
achieved is through the collective bargaining process, where employers and trade unions
negotiate on pay and terms and conditions of employment for specified groups of employees
(Freeman and Medoff, 1984; Hirsch, 2004; Turner, 2012). Overall, workers covered by
collective bargaining arrangements tend to have higher wages, lower levels of wage
inequality, better non-wage benefits, better seniority protection and better grievance systems
and lower quit rates (Freeman and Medoff, 1984).
Immigrants and the Irish labour force
Immigration has been a relatively new feature in the Irish labour force. Inward flows
occurred in a period of unprecedented growth in the Irish labour market. Between 1996 and
2006 the overall labour force increased by almost 50 per cent from 1.3 to 1.9 million (CSO,
2006) during the time period referred to as the ‘Celtic Tiger’. Initially this growth in jobs was
filled by increased participation of women in the labour market, but latterly immigrant
workers became an increasingly important source in the expansion of the labour market.
Indeed, positive net migration accounts for 60 percent of the significant increase in the
population between 2002 and 2006 (CSO, 2007). Figures for the allocation of PPS numbers
(personal public service numbers, which are required to work in Ireland) to immigrants
between 2002 and 2007 indicate that over 390,000 were from Poland alone, making the
Polish group the largest group of immigrants (apart from UK immigrants) (CSO, 2008).
As noted by Dundon et al. (2007:502) two contrasting images of foreign workers in
Ireland exist; the first image is that of highly skilled foreign workers who are central to
Ireland’s economy and work in the information technology and computer software industries,
and the second is of non-Irish national workers who are viewed as ‘a source of cheap labour,
easily disposable and found in the tertiary labour market’. Paradoxically, overall, immigrants
have higher mean levels of education compared to native workers (Cross and Turner, 2007).
Over 26 per cent of natives report reaching only either primary or lower second level,
compared to 12 per cent of immigrants, while 28 per cent of immigrants have third level
qualifications compared to 20 per cent of natives (Turner, 2010). Nevertheless, immigrants
particularly from the accession states tend to be predominantly employed in relatively low
skill occupations in the private sector and when compared to native Irish workers are underrepresented in the high skill occupations such as managers, professionals and associate
professionals and over-represented in craft, personal services, plant operatives and labouring
jobs (Fitzgerald, 2006: 19). Immigrant workers are twice as likely to be covered by a Joint
Labour Committee (JLCs) accounting for 25 per cent of all workers covered (Turner and
O’Sullivan, forthcoming). JLCs are tripartite statutory bodies with employer and union
representatives that set minimum legally binding wage rates and conditions for workers
where collective bargaining is poorly developed and pay relatively low such as in hotels,
catering, security, contract cleaning and retail sectors.
What unions can do for immigrant workers
There is an explicit premise in much of the literature that unions are good for immigrants and
can deliver significant benefits (see Milkman, 2007; Tillie, 2004; Fitzgerald and Hardy,
2010). Evidence from the US indicates that union representation substantially improves the
pay and benefits received by immigrants (Schmitt, 2010). More significantly perhaps
unionisation has the biggest impact on the wages and benefits of immigrant workers in the 15
lowest-wage occupations, raising wages by almost 20 per cent and more than doubling health
and retirement plan coverage rates (Schmitt, 2010). Beyond the immediate instrumental and
material benefits of union membership it can be argued that unions provide wider social
advantages to immigrant workers. Being a member of a trade union can strengthen the role of
the workplace as a mechanism for the social integration of immigrants into the host country.
When immigrants secure employment and start to participate in the work life of the host
society then social integration and community involvement are likely to follow (Borjas, 1995;
Putnam, 2000). Immigrants can establish social relationships with indigenous locals at the
workplace facilitating cultural and economic integration (Valenta, 2008). Thus the workplace
is crucial to the overall integration process of immigrants and exclusion from work is a
source of more general exclusion from society. Union membership appears to increase
immigrant social networks and individual social capital and is associated with higher levels
of political participation (Tillie, 2004). However in this paper the focus is primarily on the
instrumental benefits of union membership and representation.
Why are immigrant workers reluctant union joiners?
Despite the instrumental advantages of union representation and collective bargaining
immigrant workers appear reluctant to join trade unions. A national level survey in 2005
indicated that Irish workers are almost three times more likely to be union members than their
immigrant counterparts (Turner et al, 2008). At 14 per cent, union density among immigrants
compares poorly with 37 per cent of Irish workers. A survey of Polish immigrant (the largest
single non-Irish national group in Ireland) indicated that only 8 per cent of respondents
reported being a member of a trade union (Turner et al, 2008). A number of factors suggest
possible differences in union availability and union joining for immigrant workers. Union
availability is extensive in the public sector where unions are accorded a high level of
legitimacy and opposition is negligible; more extensive in industry than services and in large
firms than small firms. Immigrant workers have less opportunity to acquire union jobs due to
their limited access to highly unionised public sector jobs (Defreitas, 1993:299). Immigrants
are more likely to work in low skilled jobs in the services sector and in smaller firms in the
retail and construction sectors, contributing to lower unionisation levels in these sectors
(Grünell and van het Kaar, 2003). Consequently, immigrant workers are less likely than Irish
nationals to work in organisations with a union presence. Hence union availability is likely to
be lower for immigrant workers than Irish nationals.
Aside from the limited access to unionised sectors there are also a number of
additional reasons to expect immigrant workers will have a lower propensity to join a union.
Even where a union is available immigrant workers may not be aware of its existence
because of language difficulties and limited social contacts in the workplace (Howe, 2004;
Fang and Heywood, 2006). Moreover, because of their marginalisation, immigrant workers
are likely to be vulnerable to employer pressure not to join and less likely to speak out against
employers for any injustices or unfair treatment for fear of employer retaliation (Dundon et
al., 2007; Ness, 1998; Holgate, 2005). Immigrants often find themselves in secondary labour
markets characterised by low skills and wages. In this market the right ‘attitudes’ rather than
‘skills’ are the important qualities that employers want. Immigrants provide willing
subordinates, a good work ethic and know their place (Waldinger and Lichter, 2003: 36-41).
These factors can be compounded by a lack of understanding about Irish trade unions and
how they work. In low skill occupations in the private sector and smaller firms, these factors
are likely to be more pronounced as immigrants are more open to employer scrutiny.
In terms of individual characteristics, immigrants are generally young and young
workers are often perceived to be less likely to join unions. Immigrant workers are unlikely to
be part of any social network where the ‘reputation’ effect from being a union member has
any meaning. Unless informed, immigrants will not be cognisant of any social custom or
‘norm’ regarding union membership. Similarly, immigrants are likely to have little sense of
solidarity with Irish workers or have any ideological affinity with Irish trade unions. A
substantial proportion of immigrants usually expect to return home. With temporary migrant
labour, jobs are merely a means to an end – to save and send money back home, the true
location of their social identity (Piore, 1979).
In addition to the position and characteristics of individuals, trade union attitudes and
policies with regard to migrants may be an important factor. It has been argued in other
national contexts, that trade unions’ slow response to structural changes has contributed to the
lack of membership from non-standard employments (Burgess, 2000). Kahmann (2006: 186)
is less sanguine, arguing that ‘trade union responses to labour migration can be situated on a
continuum ranging from exclusion to inclusion, expressing a tension inherent in unionism on
the question of labour migration’. However, others believe that unions have not pursued a
restrictive approach and have focused more on organising migrants (Haus, 2002; Milkman,
2006; Watts, 2002). In Ireland, Dundon et al. (2007) suggest that union responses to
migration have differed somewhat depending on the type of migrant. They suggest that union
bargaining over wages and conditions was generally confined to EU/EEA immigrant workers
while union activities with regard to non-EU/EEA immigrant workers concerns were limited
to non-traditional union concerns. In this regard, unions engaged in ‘soft organising’ - with
attempts to make unions attractive to migrants, initiating anti-discrimination campaigns,
providing free information on employment rights and language training. However,
immigrants have seemingly not been included in union structures and Dundon et al. (2007:
516) argue that union efforts had not had ‘a significant impact on the membership of any of
the unions interviewed’.
For those immigrant workers that are unionised, this paper examines whether membership is
of benefit to them both in comparison to non-unionised immigrants and unionised Irish
workers. Specifically, we address the following questions:
1. The extent of collective bargaining coverage levels of immigrant workers compared
to Irish nationals in the private sector?
2. Does union representation benefit unionized immigrant workers compared to nonunion immigrant workers?
3. Does union representation deliver the same benefits for unionized immigrant workers
as for comparable Irish national workers?
4. Are there differences in the benefits of representation for low paid and high paid
immigrant workers?
Methodology
The National Employment Survey (NES) is a workplace and earnings survey by the Central
Statistics Office (CSO). The reference month for the 2008 NES was October 2008. The
survey provides measures of individual characteristics such as union membership, collective
bargaining coverage, sector, occupation, age, sex and educational attainment. It provides data
on individual employee earnings including overtime and shift allowances, together with
weekly hours worked. Public sector employees are excluded from our analysis for two
reasons. First the majority of employees in the public sector are union members and second,
collective agreements tend to extend universally to all employees in the public sector. The
particular benefit of the NES is that it is a matched employer-employee survey. The employer
completes a questionnaire with basic organisational details and practices and certain payrolltype details for the sample of employees. Each employee in the sample chosen completed a
questionnaire, providing information on age, gender, educational attainment, nationality,
length of time in paid employment and other job-related characteristics (CSO, 2011).
In total 9000 enterprises were sampled and almost 5000 enterprises responded - a
response rate of over 50 per cent. Almost 100,000 employees from these enterprises were
sampled and 65,535 completed the questionnaire – a response rate of over 60 per cent. In
total 22 per cent (14,619) of respondents worked in the public sector and 78 per cent (50,916)
in the private sector. To ensure that the NES is representative of the national labour force, a
comparison is made with the National Quarterly Household Survey (CSO, Standard Report
on Methods and Quality for NES) and a survey weight is provided by CSO that allows the
NES to be grossed up to the employed labour force of approximately 1.6 million employees.
The main measures used in this paper are those relating to union membership, coverage of the
firm by a collective agreement, hourly earnings and pension and healthcare benefits provided
by the employer.
Results
Table 1 outlines the labour market characteristics separately of all host country nationals and
non-Irish (immigrants) in the private sector of the economy. Focusing first on Irish nationals
and gender we can see that males account for a higher proportion of employees (54%) in the
private sector and a higher proportion of union members (57%) than females (columns 1 and
2). Union members tend to be over represented in the older age categories. Workers in the 46
to 55+ categories comprise 27 per cent of employees but 39 per cent of union membership.
Lower levels of education and working in industry tend to be more closely associated with
union membership. Unionised workers are less likely to have a third level qualification, 33
per cent compared to 41 for the overall population and more likely to have a second level
education or less, 66 per cent compared to 57 per cent. Professional workers make up 29 per
cent of employees but only 18 per cent of union members. Alternatively, service/manual
workers account for almost half (48%) of union members and 42 per cent of the population.
A minority of workers in the private sector (31%) are covered by a collective agreement
while 69 per cent of union members are covered.
Turning then to the labour market characteristics of non-Irish workers in the private
sector, there are similarities and differences with national workers (columns 3 and 4).
Principle differences are with age, education and occupation. A majority of immigrant
workers (53%) are in the younger, 25 to 35 age category. A substantially higher proportion of
immigrant workers, 32 per cent compared to 19 per cent of nationals have been educated to
degree or higher level. These higher educated immigrants make up 27 per cent of unionised
immigrants compared to 11 per cent of unionised nationals. Yet immigrant workers are more
likely to work in service/manual jobs, 56 per cent compared to 42 per cent of nationals, and a
higher proportion of these workers account for union membership, 59 per cent compared to
48 per cent of nationals.
When we examine union membership, overall, 24 per cent of all nationals in the
private sector are union members compared to 14 per cent of immigrant workers (columns 5
and 6). Union density levels for nationals are relatively higher for workers over 46 years of
age, with lower levels of education, employed in industry and in white collar/skilled and
service/manual occupations. Union density levels for non-Irish workers are relatively higher
among older workers (though there are small numbers in this category), among the lower
educated, those employed in industry and health and education. However, there appears to be
only a weak relationship between working in white collar/skilled and service/manual
occupations and the propensity to join a union.
Table 1: Characteristics of Private sector all employees, Immigrants and union density
rates, NES 2008
Private sector
Private sector
Union density rate
Irish national
Non-Irish
workers
employees
(excluding
immigrants)
All
Union
All
Union
All
Immigrants
Density all
/
/
/
/
24
14
Male
Female
54
46
57
43
56
44
58
42
25
23
15
13
Under 25
25-35
36-45
46-55
55+
16
33
24
17
10
9
27
25
26
13
14
53
22
8
3
12
53
20
11
4
14
20
25
36
32
12
14
13
19
22
Primary/none
High second
Tech/Diploma
Degree+
25
32
24
19
32
34
23
11
14
24
30
32
20
26
27
27
31
25
24
13
19
15
13
12
Industry
Construct/tran
s
Accomm/Reta
il
Health/educ
Rest
32
29
37
24
27
34
37
25
28
20
19
10
6
7
10
7
25
9
11
22
12
20
11
19
13
18
26
/
17
/
Full time
Part time
78
22
82
18
85
15
83
17
25
20
14
16
Professional
WC/Skilled
Service/Manu
al
29
29
42
18
34
48
22
23
56
17
24
59
15
28
28
11
15
15
Covered CA
Not covered
CA
31
69
69
31
25
75
65
35
/
/
/
/
26003
997122
186554
N (grossed up 997122 238315
186554
figures)
Appropriate weights applied
Source: National Employment Survey 2008.
In table 2 the raw mean hourly earnings for union and non-union nationals and non-Irish are
compared. The mean hourly earnings of both national and non-Irish union members appear to
be larger albeit moderately than non-union workers. Both national and non-Irish workers
covered by a collective agreement enjoy even higher mean earnings than that provided by
union membership. As table 1 indicated only around two thirds of union members are
covered by a collective agreement and in Ireland union members not covered are generally
excluded from any of the benefits. Compared to unionised and covered workers wage
dispersion levels as measured by the standard deviation and variance are higher among nonunion workers and those not covered by a collective agreement. Thus wage inequality tends
to be higher among non-union workers and this is also the case for non-Irish workers. Yet the
mean hourly earnings for immigrant workers are considerably lower than Irish nationals.
Although immigrant workers appear to benefit from unionisation compared to non-union
immigrants their mean hourly earnings fall well below that of even non-union nationals.
Table 2: Mean hourly earnings by unionisation and collective agreement coverage
(using earnings1)
Private sector
Private sector Non-Irish employees
Irish national workers (excluding
by unionisation (N=186554)
immigrants)
Average hourly earnings
N=997122
Union
Nonunion
Covered
by CA
Not
covered
Union
Nonunion
Covered
by CA
Not
Covered
Mean
€20.0
€19.4
€21.7
€18.6
€16.9
€16.1
€17.1
€15.9
St. Dev.
9.96
13.7
12.6
12.9
8.7
10.2
9.7
10.1
167
76
104
94
102
Variance
99
188
159
Appropriate weights applied
Source: National Employment Survey 2008.
Given the skewed asymmetrical nature of the earnings distribution in our sample (see
appendix) the dependent variable earnings per hour is best transformed into a logarithm of the
original variable to correct the distribution to a normal one. Regressing unionisation on the
log of hourly wage earnings indicates that national union members have an average union
premium of 9 per cent compared to non-union workers and this rises to 19 per cent for
workers covered by a collective agreement (table 3). Both unionised and covered immigrant
workers receive an hourly wage premium of 8 per cent. However such wage premiums
possibly reflect differences in human capital attributes such as education and skills and
employment factors such as firm size and sector. Controlling for these factors substantially
reduces the hourly earnings premiums for unionised for Irish and immigrant workers (table
3). The wage premium for national unionised workers decreases to 5 per cent and even more
dramatically to 6 per cent for workers covered by a collective agreement. For unionised
immigrants the decrease in the hourly earnings premium is less substantial falling to 6 per
cent for union membership and 4 per cent for coverage. Nevertheless union membership
appears to deliver a modest wage premium of a relatively similar magnitude to both nationals
and immigrant workers.
Table 3: Wage premiums by unionisation and collective agreement coverage (using
earnings=sizeearns1)
Private sector
Private sector Non-Irish employees
Irish national workers (excluding
by unionisation (N=186554)
immigrants)
Average hourly earnings
N=997122
Union
Mean log
wagea
Wage
premiuma
F scored
N
(unweighted)
2.94
Average hourly earnings
NonCovered
Not
union
covered
2.88
3.0
2.84
Union
Nonunion
Covered
Not
Covered
2.8
2.7
2.8
2.7
+9%b
+19%
+8%
+8%
107***
1061***
667***
1085***
431349
43139
7762
With controlsc
Wage
premium
F score
N
(unweighted)
7762
With controls
+5%
+6%
+6%
+4%
2926**
*
2941***
461***
463***
41105
41105
7502
7502
* F<0.05 ** F<0.01 ***F<0.001
a
The independent variables union membership and collective agreement coverage were regressed on the
dependent variable hourly log earnings (Weighted data used).
b
The percentage difference is calculated from the exponential of the unstandardized regression coefficient for
the variables union membership and collective agreement coverage in each equation. The F score for each
equation is also given.
c
The following controls were introduced into the regression equations: Gender, age, education, full time or parttime status, occupation, firm size, industrial sector, years of service (see table ??).
d
F scores and Ns are un-weighted
Source: National Employment Survey 2008.
As noted above union members tend to have lower levels of wage dispersion than non-union
workers. Unionised Irish members are less likely to be in the low pay category than nonunion nationals, 16 per cent compared to 27 per cent. Conversely, 51 per cent of national
unionised workers have average mean hourly earnings above the median compared to 40 per
cent of non-union nationals. A greater number of non-Irish compared to Irish national
workers are in the low pay category and a smaller number have above the median hourly
earnings. Even so unionisation makes a difference: 37 per cent of non-union immigrants are
low paid compared to 25 per cent of unionised workers and 31 per cent of unionised
immigrants compared to 25 per cent of non-union workers earn above the median.
In terms of working hours, unionised workers, both Irish nationals and particularly
immigrants, are more likely to work shifts (based on the employee survey). However, only 10
per cent of unionised nationals and 12 per cent of unionised immigrant workers receive a
shift allowance (based on the employer survey). Thus it appears that 29 per cent of nationals
and 41 per cent of unionised immigrants are unpaid for shift work. Although non-union
national and immigrant workers are less likely to work shifts they are also proportionately
less likely to get paid for working shifts. Immigrant workers on average work longer hours
per week but there is little difference between union and non-union workers. Unionised
workers particularly immigrants are more likely to work overtime than non-union workers. It
may be the case that non-union workers receive only a flat rather than a premium rate of pay
for extra working hours but it is not possible to verify this from the present survey.
Apart from a wage premium much of the evidence on the advantages of union
membership show that unionised workers tend to have higher levels of benefit coverage in
areas such as pensions than non-union workers. The data reveals that unionised national
workers in the private sector are more likely to be covered by a pension scheme than nonunion workers, 80 per cent compared to 58 per cent (table 4). Unionised immigrants are also
more likely to be covered by a pension scheme though at a lower level than unionised
nationals, 62 per cent compared to a 47 per cent of non-union immigrants. Health assurance
coverage is relatively similar for both union and non-union national and immigrant workers.
Table 4: Unionisation by pensions and Health insurance
Private sector Irish
nationals excluding
immigrants (N=997122)
Union
%
Pensionsa
Health Assurance
80
31
Nonunion
%
58
29
Low paid
Up to median
Above Median
16
33
51
Work shift work
Get paid for shift work
Mean hours worked per week
Mean hours overtime per week
Private Non-Irish
employees sector by
unionisation
(N=186554)
Union
Non-union
%
%
62
32
47
31
27
33
40
25
44
31
37
38
25
39
10
18
3
53
12
39
5
34
1.3
33
0.4
36
1.96
36
0.9
a
Respondents were asked: ‘Does your employer offer you (a) Pension schemes and (b) Health Assurance.
Responses are scored 1=Yes; 0=No.
Appropriate weights applied
Source: National Employment Survey 2008.
To further test the relationship between union membership and instrumental benefits table 5
uses multivariate analysis to compare the outcomes for union membership for Irish nationals
and immigrant workers. This allows us to control for the possible effects of individual human
capital attributes such as education, occupation and experience and also structural
characteristics such as firm size and sector. Unionised Irish nationals are 1.8 times more
likely to earn above the median and 1.3 times more likely to earn up to the median than nonunion workers (column 1). They are also more likely to work overtime, shift work and be
covered by a pension scheme (1.8 times) but are less likely to have employer paid health
assurance. A similar pattern emerges from a comparison of unionised immigrant workers
with non-union immigrants (column 2). Unionised immigrants are twice as likely to earn
above the median hourly pay, 1.5 times more likely to earn up to the median, work more
overtime hours, shift work, get paid for shift work, have wider pension coverage but less
likely to have paid health assurance. Column 3 compares unionised Irish nationals with
unionised non-national workers. Unionised nationals are twice as likely to receive above the
median hourly earnings, twice as likely to be covered by a pension scheme but less likely to
work overtime hours and shifts. Thus unionised nationals appear to enjoy greater benefits
from membership than non-nationals. Column 4 compares non-union Irish nationals with
unionised non-nationals. Non-union Irish nationals are almost three times less likely to work
overtime and 4 times less likely to work shift work than unionised immigrants but are 1.8
times more likely to be have employer paid health assurance. However there is little
difference in the substantive areas of earnings, pension coverage and paid shift work.
Finally column 5 examines whether unionised non-nationals in the low pay category fare
differently from those with above the median earnings when controlling for all relevant
variables. Unionised immigrants in the high earnings category are slightly more likely to
work overtime hours, 2.6 times more likely to get paid for shift work and 4.8 times more
likely to be covered by a pension scheme. Thus it appears that compared to union members in
the low pay category unionised immigrants with above median earnings appear to enjoy more
benefits. Some caution however is required as the numbers in the regression equation are
relatively small and standard errors substantial.
Table 5: Union membership and comparative outcomes (unweighted) a
Private sector Private sector
Unionised
Non-union Nats
Irish
Non-Irish
nationals and
and unionised
nationals
nationals
non-Irish
non-Irish
1
2
3
4
Union=1
Union=1
Union Ir=1
Nonunion Ir=1
Non-union=0 Non-union=0
Union NI=0
Union NI=0
Above Median
1.8***
2.2***
2.0***
1.1 (ns)
hourly earnings
(0.05)
(0.1)
(0.1)
(0.1)
Up to median
1.3***
1.5***
1.2 (ns)
-1.0 (ns)
earnings
(0.05)
(0.01
(0.1)
(0.1)
Ref category: Low
pay
Mean hours
worked per week
Mean hours
overtime per week
Work shift work
Get paid for shift
work
Pensions
Health Assurance
% correct
Nagelkerke R2
Chi.sq
N
Unionised nonnationals: High
pay v low pay
5
High pay=1
Low pay=0
/
/
1.0 (ns)
(0.06)
1.8***
(0.04)
3.5***
(0.04)
0.99 (ns)
(0.05)
1.0 (ns)
(0.09)
1.9***
(0.08)
1.9***
(0.08)
1.4***
(0.1)
1.0 (ns)
(0.01)
-1.8***
(0.08)
-1.3***
(0.08)
-0.9 (ns)
(0.1)
-1.0 (ns)
(0.01)
-2.9***
(0.08)
-3.9***
(0.08)
-0.8 (ns)
(0.1)
0.95**
(0.01)
1.3***
(0.05)
-0.9(ns)
(0.2)
2.6***
(0.3)
1.8***
(0.04)
-2***
(0.03)
1.4***
(0.08)
-1.7***
(0.09)
2.1***
(0.09)
-1.2 (ns)
(0.08)
1.2 (ns)
(0.08)
1.8***
(0.08)
4.8***
(0.27)
-0.7 (ns)
(0.3)
81
0.32
8860***
37309
85
0.17
685***
6876
90
0.23
1136***
9523
96
0.22
1763***
29878
83
0.58
347***
626
* P<0.05 ** P<0.01 ***P<0.001
a
The following controls were introduced into the four regression equations: Gender, age, education, full time or
part-time status, occupation, firm size, industrial sector, years of service .
Source: National Employment Survey 2008.
Discussion and Conclusion
Using a robust national matched sample of employee/employer responses this paper
examined four questions regarding the unionisation of immigrant workers in the private
sector. First we examined the unionisation rates of immigrant workers compared to Irish
workers. In line with research internationally, we find that unionisation rates are lower for
immigrant workers than Irish workers, highlighting a lack of focus on recruiting immigrant
workers by trade unions in Ireland. However, we also see that certain individual and work
characteristics attract higher unionisation rates for both immigrant and Irish workers e.g. for
males, older workers, lower education levels, non-professional occupations, industry and
health/education sectors. The only category where immigrant and Irish national workers
differ is in relation to working hours.
Second, we addressed whether union membership and representation benefits
unionized immigrant workers compared to non-union immigrant workers. The instrumental
nature of union membership in terms of wage premium is confirmed here. Union membership
appears to deliver a modest wage premium of a relatively similar magnitude to both nationals
and immigrant workers. Wage inequality tends to be higher among all non-union workers
including immigrant workers than unionised Irish nationals and non-nationals. Unionised
immigrants are twice as likely to earn above the median hourly earnings, be paid for shift
work and have greater pension coverage than non-union immigrants. Consequently it can
plausibly be argued that union membership provides moderate instrumental benefits for
immigrant workers. However the small size of the earnings premium may be some concern as
one of the main identifiable benefits for recruiting new members is the wage premium
afforded to union members. While union membership is attractive to the extent that it is
instrumental in achieving material goals it is also the case that a significant reason for joining
a union is protection against unfair treatment and arbitrary management actions (Waddington
and Whitson, 1997).
A third question concerned whether union membership delivered the same benefits
for unionized immigrant workers as for comparable unionized Irish nationals. Our results
indicate that while immigrant workers benefit from unionisation compared to non-union
immigrants, their mean hourly earnings fall well below that of even non-union nationals.
Moreover though unionised immigrants are more likely to be covered by a pension scheme it
is significantly lower than unionised nationals. Thus unionised nationals appear to enjoy
greater benefits from membership than non-nationals. However a comparison of non-union
Irish nationals with unionised immigrant workers indicated little difference in the substantive
areas of earnings, pension coverage and paid shift work between the two groups. Lastly, the
question of possible differences in the benefits of representation for low paid and high paid
immigrant workers was examined. Results indicate that compared to union members in the
low pay category unionised immigrants with above median earnings seem to enjoy more
benefits.
Implications
There are a number of implications we can derive from the data. Firstly, immigrants are an
untapped source of new members for those charged with union leadership. Given that
immigrants did not leave Ireland in the face of the most severe economic recession in decades
(CSO, 2011, Krings, 2009) and decided to remain in Ireland, they can be viewed as a longterm source of new membership if specifically targeted by union organising campaigns. For
unions there is some scope for optimism as union membership levels among immigrant
workers tends to increase with age which can be viewed as a proxy for length of time in the
country. Indeed, studies generally indicate that immigrant integration into host country
networks increases with time. Length of residence in the country is likely to improve the
possibility that immigrants develop a closer affinity with local movements such as trade
unions. It may well be that immigrant workers over time have a greater propensity to join a
union because of increased contact with a peer network that includes (older) Irish workers.
While immigrants’ workers do benefit from union membership, their earnings and
benefits however, still lag behind Irish-national workers. Being a union member alone does
not guarantee higher wages. Over a third of unionised immigrants are in workplaces that do
not recognise or negotiate with unions and are therefore not covered by a collective
agreement. In such a scenario union membership may help immigrants in terms of
information and advice but these workers cannot benefit from any wage premium or
improved conditions associated with a collective agreement.
Of course the data does not provide a full picture of the potential benefits of unions
for immigrants. Unions also provide a collective voice for workers – ‘a sword of justice’
providing information and representation in individual and collective matters at work and
crucially, protection against arbitrary management treatment. However, the discrepancy
between average mean hourly earnings between Irish-nationals and immigrants is
problematic for both employees and unions. For unions, it raises questions as to how one
section of their membership (immigrants) appear not to benefit in the same extent as
unionised Irish-nationals. Unions may find it difficult to attract immigrants into the union if
there is a manifest differential in pay and conditions with Irish-national workers. Reducing
pay inequality will require stronger organisation of workers, a stronger focus on migrant
concerns in union agendas and particularly in collective bargaining, and effective
enforcement by state bodies of minimum pay rates.
References
Blanchflower, D.G. and A. Bryson (2004) 'What effect do unions have on wages now and
would Freeman and Medoff be surprised?', Journal of Labor Research, XXV(3), Summer,
pp. 383-414.
Blanchflower, D.G. and A. Bryson (2003),'Changes over time in union relative wage effects
in the UK and the US revisited,' in J.T. Addison and C. Schnabel, eds. International
handbook of trade unions. Cheltenham, UK: Edward Elgar, pp. 197–245.
Borjas, George, J. (1994) “The Economics of Immigration”. Journal of Economic Literature
32, 4: 1667-1717.
Budd, J. (2005) ‘Employment with a Human Face: The Author Responds’, Employee
Responsibilities and Rights Journal, Vol. 17, No.3: 191-199.
Burgess, J. (2000) ‘Globalization, non-standard employment and Australian trade unions’,
Asia Pacific Business Review, 6(3-4), 93-113.
Callan, T. and Reilly, B. (1993) ‘Unions and the Wage Distribution in Ireland’, The
Economic and Social Review, Vol.24, No.4: 297-312.
Cross, C. and Turner, T. (2007) “Irish Workers Perceptions of the
Impact of Immigrants: A Cause for Concern?” Irish Journal of Management 27, 2: 215-245.
CSO (2006) Quarterly National Household Survey: Quarter 4, 2005. Cork: Central Statistics
Office.
CSO (2007), Foreign Nationals: PPSN Allocations and Employment, 2002-2006, December
release, Central Statistics Office, Cork.
CSO (2008), Census 2006: Non-Irish Nationals Living in Ireland, Central Statistics Office,
Cork.
CSO (2009) Quarterly National Household Survey, Quarter 2 2009 (September), Cork:
Central Statistics Office.
CSO (2011) National Employment Survey 2008 and 2009. Dublin: Stationery Office.
Cranford, C., Vosko, L.F and Zukewich, N. (2003) ‘Precarious Employment In The Canadian
Labour Market: A Statistical Portrait’, Just Labour, 3: 6-22.
Crouch, C. (1982) Trade Unions: The Logic of Collective Action. Fontana Paperbacks.
D’Art et al, (2013) ‘Introduction’ in Tom Turner, T., D’Art, D. and O’Sullivan, M. (eds –
forthcoming) Are Trade Unions still Relevant? Union Recognition 100 Years On. Dublin:
Orphen Press.
Defreitas, G. (1993) Unionization Among Racial and Ethnic Minorities, Industrial and Labor
Relations Review, 46(2): 284-301.
Dundon, T., González-Pérez, M.A. and McDonough, T. (2007) ‘Bitten by the Celtic Tiger:
Immigrant Workers and Industrial Relations in the New `Glocalized' Ireland’, Economic and
Industrial Democracy, 28(4): 501-522.
European Commission (2011) Industrial Relations in Europe 2010, available at:
http://bit.ly/kcid8h, accessed 13/02/2013.
Fang, T. and Heywood, J. Immigration, Ethnic Wage Differentials and Output Pay:
Canadian Evidence from the WES, Paper presented at British Journal of Industrial Relations
Workshop, 2006.
Fang, T. and Verma, A. (2002) ‘Union Wage Premium, Perspectives on Labour and Income
(Statistics Canada, catalogue no. 75-001-XIE) 3, no. 9. September 2002, online edition.
Finch, Tim, Latorre, Maria C., Pollard, Naomi and Rutter, Jill (2009) “Shall We Stay or Shall
We Go? Re-migration Trends among Britain’s Immigrants”. Institute for Public Policy
Research, London.
Fitz Gerald, J. (2006) The Labour Force Needs of a Cosmopolitan Economy, 14th Annual
John Lovett Memorial Lecture, March, University of Limerick.
Fitzgerald, I. and Hardy, J. (2010) ‘‘Thinking outside the box’? trade union organizing
strategies and polish migrant workers in the United Kingdom’, British Journal of Industrial
Relations, 48(1), 131-150.
Flanders, A. (1970) Management and Unions: Theory and Reform of Industrial Relations.
London: Faber and Faber.
Freeman, R. (1994) ‘American Exceptionalism in the Labor Market: Union/Nonunion
Differentials in the United States and Other Countries’ in in Kerr, C. and Staudohar, P. (eds.)
Labor Economics and Industrial Relations: Markets and Institutions. Cambridge, MA:
Harvard University Press.
Freeman, R. and Medoff, J. (1984) What Do Unions Do? New York: Basic Books.
Grünell, M. van het Kaar, R. (2003) Migration and industrial relations. Available at
http://www.eiro.eurofound.eu.int/2003/03/word/at0212203s.doc, Accessed 21 August 2006.
Hartley, J.(1992) ‘Joining a Trade Union’ in Hartley, J. and G. Stephenson (eds.)
Employment Relations, Oxford: Blackwell.
Haus, L.A. (2002) Unions, Immigration, and Internationalization: NewChallenges and
Changing Coalitions in the United States and France. New York: Palgrave Macmillan.
Hirsch, B. (2004) ‘Reconsidering Union Wage Effects: Surveying New Evidence on an Old
Topic’, Journal of Labor Research, 25(2): 233-266.
Hirsch, Barry T. and Edward J. Schumacher (2004) “Match Bias in Wage Gap Estimates Due
to Earnings Imputation.” Journal of Labor Economics 22 (July)
Hix, Simon and Noury, Abdul (2007) “Politics, Not Economic Interests: Determinants of
Migration Policies in the European Union”. International Migration Review 41, 1 March
2007: 182–205.
Holgate, J. (2005) ‘Organising Migrant Workers: A Case Study of Working Conditions and
Unionization in a London Sandwich Factory’, Work, Employment and Society, Vol. 19,
No.3:463-480.
Howe, D. (2004) “British trade unions should organise a grand mobilisation to support
migrant workers” New Statesman, 133(4675):13-13.
Kahmann, M. (2006) ‘The Posting of Workers in the German Construction
Industry: Responses and Problems of Trade Union Action’, Transfer, 12(2):
183–96.
Krings, T. (2009) ‘A Race to the Bottom? Trade Unions, EU Enlargement and the Free
Movement of Labour’, European Journal of Industrial Relations, 15(1): 49-69.
McKay, S., Clark, N. and Paraskevopoulou, A. (2011) Precarious Work in Europe Causes
and consequences for the Agriculture,Food and Tourism sectors. European Federation of
Food, Agriculture and Tourism Trade Union.
Metcalf, D. (1982) ‘Unions and the Distribution of Earnings, British Journal of Industrial
Relations, Vol.16, No.2: 163-169.
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’, National Institute
Economic Review, No. 176, April: 61-75.
Milkman, R. (2006) L.A. Story: Immigrant Workers and the Future of the U.S.
Labour Movement. New York: Russell Sage Foundation.
Ness, I. (1998) “Organising Immigrant Communities: UNITE’s Worker Center Strategy” in
Bronfenbrenner, K., Friedman, S., Hurd, R.W., Oswald, R.A. and Seeber, R.L. (eds)
Organising to Win. New York: Cornell University Press.
OECD (2003) Labour Shortages and the Need for Immigrants: A Review of Recent Studies,
Part II. Accessed Januray 28th 2012. www.oecd.org/dataoecd/27/43/31857112.pdf
Piore, M. (1979) Birds of Passage: Migrant Labor and Industrial Societies. London:
Cambridge University Press.
Putnam R (2000) Bowling Alone: The Collapse and Revival of American Community, New
York, NY: Simon and Schuster.
Sayer, A. (2000) ‘Moral Economy and Political Economy’, Studies in Political Economy,
Vol.61: 79-104.
Schmitt, J. (2010) Unions and Upward Mobility for Immigrant Workers. Center for
Economic and Policy Research, March, Washington. www.cepr.net.
Schmitt, J. and Mitukiewicz, A. (2011) Politics Matter Changes in Unionization Rates in
Rich Countries, 1960-2010. Washington: Center for Economic and Policy Research.
Tillie, J. (2004) ‘Social Capital of Organisations and their Members: Explaining the Political
Integration of Immigrants in Amsterdam’, Journal of Ethnic and Migration Studies, Vol.3,
No.3: 529-541.
Turner, T. (2012) In Pursuit of Members interests: Trade Unions, Wage Premiums and
earnings dispersion in the Irish private sector. paper presented at Lisbon conference.
Turner, T. (2010) “The Jobs Immigrants Do: Issues of Marginalisation in the Irish Labour
Market”. Work, Employment and Society, 24, 2: 318-336.
Turner, T. and O’Sullivan, M. (forthcoming 2013) ‘Economic crisis and the restructuring of
wage setting mechanisms for vulnerable workers in Ireland’, Economic and Social Review.
Turner, T., O'Sullivan, M. and D'Art, D. (2011) ‘Organising methods and member
recruitment in Irish trade unions’, Employee Relations, Vol. 33, No. 5: 516 – 531.
Turner, T., D’Art, D. and O’Sullivan, M. (2008) ‘Union Availability, Union Membership and
Immigrant Workers: An Empirical Investigation of the Irish Case’, Employee Relations,
Vol.30, No.5:479-493.
Turner, Bryan S. (2007) “The Enclave Society: Towards a Sociology of Immobility”.
European Journal of Social Theory 10, 2: 287-303.
Valenta M (2008) The workplace as an arena for identity and affirmation and social
integration of immigrants. Forum Qualitative Sozialforschung / Forum: Qualitative Social
Research 9 (2), Art. 14, available at http://nbn- resolving.de/urn:nbn:de:0114-fqs0802140,
(accessed 18 November, 2010).
Waddington, J. and Whitson, C. (1997) ‘Why Do People Join Unions in a Period of
Membership Decline, British Journal of Industrial Relations, vol.35, 4: 515-546.
Waldinger, R. and Lichter, m. (2003) How the Other Half Lives: Immigration and the Social
Organisation of Labour. Berkeley: University of California Press.
Watts, J. (2002) Immigration Policy and the Challenge of Globalization. Ithaca,
NY: Cornell University Press.