Discrimination in the labour market: nationality,ethnicity and the recession

Abstract

Previous research shows that immigrants, in common with other groups that suffer disadvantage in the labour market, are more vulnerable during recession. However, little research has focused on the impact of the Great Recession on work-related discrimination. We examine the extent to which discrimination varies across different national-ethnic groups in Ireland, and whether discrimination increased between 2004, during an economic boom, and 2010, in the midst of a severe recession. Our analysis draws on two large-scale nationally representative surveys on the experience of labour market discrimination. We find that overall non-Irish nationals do experience higher rates of work-based discrimination and that there is substantial variation in discrimination across national-ethnic groups. However we find no evidence to suggest that self-reported discrimination increased during the recession.

Keywords

discrimination ethnicity immigrants labour market recession

Introduction

A growing body of research shows that immigrants suffer multiple disadvantages in the Irish labour market, with lower employment, higher unemployment and lower wages than Irish nationals (Barrett and Duffy, 2008; O’Connell and McGinnity, 2008). The Great Recession led to a dramatic deterioration in the Irish economy after 2008. In general, immigrants are more exposed to the consequences of economic downturns, and this is also the experience in Ireland. Against this backdrop of disadvantage experienced by immigrant groups, we investigate whether immigrants are more likely to report experience of discrimination in the labour market; whether such discrimination differs by nationality and ethnicity; and whether the incidence of discrimination increases in the adverse labour market conditions of the recession.

Our analysis draws on two large-scale nationally representative surveys that collected self-reports of the experience of discrimination. The first was conducted in 2004 in the midst of an economic and employment boom accompanied by substantial inward migration, then a novel episode in Irish demography. The second was conducted in 2010 in the midst of an economic, fiscal and employment crisis of unprecedented severity. Both surveys also collect information on nationality, ethnicity and labour market and socio-demographic indicators (CSO, 2011).

Previous research shows that immigrants, in common with other groups that suffer disadvantage in the labour market, are more vulnerable to falling employment levels and rising unemployment during recession in Ireland (McGinnity et al., 2013) as elsewhere (Hoynes et al., 2012). However, little research has focused on the impact of the Great Recession on work-related discrimination. We recognize that self-reports, in common with other methods of measuring discrimination, are not without their limitations. Self-reports may be biased upwards or downwards, and are unlikely to pick up indirect discrimination (Organisation for Economic Co-operation and Development (OECD), 2013). However, though we employ robust data in our analysis, we acknowledge that reports of discrimination on their own cannot unambiguously establish the prevalence of labour market disadvantage, and we argue that the analysis of discrimination complements the results of other approaches, including field experiments and statistical analysis of ethnic penalties in labour market outcomes.

This article contributes to the literature on discrimination in a number of ways. First, it assesses the extent of discrimination experienced by non-Irish nationals in the Irish labour market. Second, it examines the extent to which the experience of discrimination varies between national-ethnic groups. This may contribute to more rigorous and nuanced approaches to the analysis of nationality and ethnicity in future Irish research. Third, this is the first article, to our knowledge, to shed light on the experience of labour market discrimination in the Great Recession, complementing previous research on objective indicators of immigrants’ experiences in tight and slack labour markets. Previous research in the UK found that religious and ethnic penalties in unemployment increased post-recession (Khattab and Johnston, 2013).

In the next section we outline the context for our research, focusing on recent trends in migration, the labour market and the economy in Ireland. We then discuss theoretical approaches that inform our analysis, and the research questions deriving from those approaches. Next we discuss our data sources and the measurement of discrimination. The results of the analysis are then presented, followed by a discussion of the findings and their implications.

Changing migration patterns and Ireland’s boom and bust

Ireland represents an interesting case because it combines large-scale immigration into a small labour market that was almost exclusively White and Irish, with a sudden and deep recession immediately following the peak of immigration.

Ireland, historically a country of net emigration, experienced significant inward migration between the mid-1990s and 2008, during a period of rapid growth in the economy and employment. The number of foreign residents increased from 224,300, or 6 per cent of the total population in 2002 to 575,600, or 12.8 per cent in 2008, before falling back to 550,400, or 12 per cent, in the wake of the economic crisis in 2012.¹ Following EU enlargement in 2004 and 2007 there was substantial immigration from the new member states (NMS)² so that by 2008, NMS nationals were the largest group of immigrants, accounting for 5.5 per cent of the total population. Nationals of the older EU states,³ including the UK, accounted for less than 4 per cent of the population, and those from the rest of the world accounted for another 3.5 per cent (O’Connell and Joyce, 2013). Accordingly, about three-quarters of all immigrants in the latter years of the last decade were Europeans, and mostly White, while about one in four were of more diverse nationality and ethnicity.

The Irish economy moved into a deep and prolonged recession in 2008, following two decades of rapid growth. The crisis was multi-dimensional, entailing the bursting of a property bubble; a banking collapse; contraction in economic activity; state fiscal crisis; and mass unemployment (O’Connell, 2013). Total employment fell by 13 per cent between the end of 2007 and 2011, but it fell by 21 per cent among non-Irish nationals. In 2012 the unemployment rate was 14.5 per cent among Irish nationals but 17.7 per cent among non-Irish nationals. Economic collapse was accompanied by substantial migratory flows, yet a modest decline in the immigrant population because substantial out-migration, mainly of NMS nationals, was offset by substantial in-migration, mostly from the same region: those who had been displaced from the collapsing sectors/occupations appear to have been replaced by others with different, more marketable skills.

Theoretical approaches to discrimination

While measuring the extent of discrimination is challenging, identifying the underlying causes or mechanisms is even more so. Much research in the area focuses on individual motivations, though these are often difficult to measure empirically (Reskin, 2003). Classical works emphasize the role of prejudice or racial animus as a key underpinning of discrimination, with negative feelings and beliefs influencing subsequent behaviour towards that group (Pager and Shepherd, 2008). Different authors have different emphases: Becker (1957/1971) postulates that some employers and economic agents have a ‘taste for discrimination’, which influences their decisions; Blumer (1958) places more emphasis on how groups compete for scarce resources; and later work in this tradition focuses on how the salience of competition may vary in different economic and cultural contexts, and between ethnic groups (Quillian, 2006). Typically, though not always, groups may experience higher rates of discrimination if they are more visibly and/or culturally different. European evidence suggests that immigrants from Sub-Saharan Africa are most likely to perceive discrimination followed by immigrants from North Africa, Latin America and Asia (see OECD, 2013). Prejudice may not always be explicit: literature from social psychology highlights how forms of racial bias that actors are not even consciously aware of, known as implicit prejudice, may influence behaviour (Al Ramiah et al., 2010).

Another variant of preference or prejudice-based discrimination places greater emphasis on ‘in-group favouritism’, a tendency to treat in-group and out-group members differently, which may be manifested in preferential treatment of the in-group in resource allocation (e.g. in recruitment decisions) (Brekke and Mastekaasa, 2008; McGinnity and Lunn, 2011; Quillian, 2006). McGinnity and Lunn (2011) argue that it is a powerful idea, as even relatively mild in-group favouritism, which does not contain an active element of aggression or negative affect or emotion, can result in substantial discrimination in the allocation of resources.

Alternative approaches challenge the notion that prejudice is at the root of discrimination. With statistical discrimination models, differential outcomes for immigrant groups are due to information problems (Phelps, 1972). Decisions result from insufficient information on the part of employers about minority groups, and this informational deficiency can be particularly acute at labour market entry (Brekke and Mastekaasa, 2008). While preference-based discrimination relies on the presence of prejudice, statistical discrimination results from employers’ lack of information about a minority group: employers use race or migration as a heuristic guide to evaluate job applicants in the absence of adequate information (Pager and Shepherd, 2008). This approach has close parallels to the rich literature in social psychology on stereotyping, where stereotypes are beliefs about a group – for example that they are lazy, intelligent, violent, etc. – that are used as a proxy for missing information. Whereas social psychologists see stereotypes as ‘faulty and inflexible generalizations’, statistical discrimination approaches see these ‘group estimates’ as a rational response to uncertainty (Pager and Shepherd, 2008). The implication of both is that discrimination may be reduced over time if employers gain more accurate information about immigrant/minority groups, and encounter such groups more often, and the groups gain labour market experience in a country.

Sociological approaches to discrimination highlight how individual-level explanations may be mediated by organizational or national factors (Reskin, 2003). Characteristics of organizations, such as personnel practices, may constrain the biasing effects of either cognitive or attitudinal biases (Reskin, 2000). One important example of this is the use of formal, rationalized procedures in an organization for recruitment and promotion. Formalization reduces individual discretion, and this may be linked to increased representation of minorities in the organization (Pager and Shepherd, 2008). While formalization does not always reduce or eliminate discrimination, formal procedures could be associated with less discrimination. Research in Ireland is limited, but suggests that formalized procedures are more common in larger firms and in certain sectors and occupations (Russell and McGinnity, 2011).

While much research on discrimination relates to decisions at the level of individuals or small groups, societal or national context is also important (Pager and Shepherd, 2008). Structural discrimination can refer to laws and cultural institutions that impose different rules on different groups. The most obvious example in Ireland is that non-Irish EU nationals can reside and work in Ireland with similar rights to Irish nationals, whereas non-EU nationals face very different regulations. Regulations also changed during the period this article examines: prior to EU enlargement in 2004, there was a significant group of immigrants from outside the EU working under the Employment Permit system with a diversity of skill levels. Following enlargement, Irish policy was to meet labour shortages from within the EU, and to reserve the Employment Permit system to meet identified skill shortages, typically in highly skilled occupations. While this restricted non-EU immigration into Ireland, it also meant that non-EU immigrants working in Ireland tended to be more highly educated than immigrants from Eastern Europe post-2004 (McGinnity et al., 2013).

Another element of research on ‘structural discrimination’ focuses on the legacies of discrimination, and how differences may relate to past policies and practices. This has most resonance in countries with a long history of immigration and past discriminatory policies (e.g. the USA), but it does alert us to one potential source of discrimination in Ireland. Immigrants who come to Ireland seeking political asylum or protection are not allowed to work while their application is being processed. In theory this period should be short, but in practice applications can be drawn out: over 30 per cent of residents in centres for asylum seekers have been in the system for more than five years (Joyce and Quinn, 2014). Some national/ethnic groups in Ireland are more closely associated with political migration, in particular Black Africans. Immigrants who have had an extended period in the asylum system and, as a consequence, been excluded from the labour market, may have suffered poorer job prospects. Even if individuals have not been so excluded from the labour market, to the extent that employers think this is the case, this may lead to unequal treatment.

Different perspectives on discrimination also differ in their understanding of the impact of social and economic context, in particular the impact of a rapid rise in the population of national/ethnic minorities associated with immigration. From an intergroup contact perspective (Blau, 1977), as the presence of minority groups in the workplace increases, workers will have more opportunities to interact with members of other racial and ethnic groups. Such interaction allows workers from all racial/national groups to acquire information about each other, and gain personal experience with them, making them less likely to indulge in racial stereotypes and biases. As employers gain information about groups this may over-ride previously held expectations. Preference or prejudice-based approaches to discrimination may be less sanguine about the impact of a growth in the proportion of immigrants in a country. Researchers focusing on anti-immigrant attitudes argue that an increase in the share of immigrants can lead to intensified perceptions of ethnic threat due to increased economic competition and identity-based cultural conflict (Quillian, 2006; Schneider, 2008).

In a deep recession, with increased competition for scarce resources, immigrants may be especially likely to be perceived as competing with members of the host society (Esses et al., 2001). Individuals may perceive more threat and competition from minorities, particularly if the economic context entails competitive conditions (Schneider, 2008). Coenders et al. (2008) found that ethnic discrimination became more widespread in periods of high immigration and when the unemployment level had risen strongly. There is some support for this argument in Ireland, where overall attitudes to immigrants in Ireland have become more negative in the period between 2004 and 2010 (McGinnity et al., 2013). However, as discussed above, there are a number of reasons why anti-immigrant sentiment may not necessarily translate into discriminatory behaviour on the part of employers.

Measuring discrimination and previous research

Most definitions regard discrimination as differential or unequal treatment of the members of a group on the basis of their group membership (Pager and Shepherd, 2008). There is a substantial body of evidence pointing to persistent inequalities between immigrant or minority groups and nationals in the labour market, though variation exists between immigrant groups. Unemployment has been shown to be consistently higher among immigrants than natives in Europe (Fleischmann and Dronkers, 2010) and in Ireland (McGinnity et al., 2013). Immigrants and other minorities also tend to be over-educated: employed at occupational levels below their skill level in the UK (Rafferty, 2012) and in Ireland (Barrett and Duffy, 2008). There is also evidence of substantial wage penalties, whereby immigrants earn less than Irish nationals, and this varies by national group (Barrett and McCarthy, 2007).

While much of the international research on unequal treatment among immigrants focuses on both nationality and ethnicity (Rafferty, 2012), most Irish research focuses on nationality (Barrett and McCarthy, 2007; McGinnity et al., 2013). Irish evidence on the impact of ethnicity is more limited, although O’Connell and McGinnity (2008) show that Black immigrants are more likely to experience unemployment and lower level occupations, even when other factors are controlled for.

Most studies of unequal labour market outcomes among immigrants take account of differences in other factors, such as gender, education and experience. The question remains as to whether the unexplained residual differences in labour market outcomes between immigrants and nationals can be attributed to discrimination. The difficulty with such residual approaches, however, is that other influential human capital differences may not be captured in the data, resulting in inaccurate, and potentially inflated, estimates of possible discrimination (Pager and Shepherd, 2008). One alternative strategy is to measure discrimination directly through field experiments. A field experiment in Ireland found that candidates with Irish names were more than twice as likely to be called to interview than candidates with clearly non-Irish names but otherwise equivalent CVs. However there were no differences within the immigrant group (McGinnity and Lunn, 2011). This method provides powerful evidence of discrimination but is limited to certain sectors and occupations, and the groups under study at a particular point in time.

Self-report studies ask respondents about their experience of discrimination and can be collected in large-scale representative surveys, which allow for comparison between the experience of minority and majority populations. This method plays an important part in tracking change and stability in discrimination over time. However, self-reports are subjective, relying on the assessment of the individual, which may vary depending on the perspective of the respondents, their expectations and the information available to them (Blank et al., 2004). Strong survey design can minimize this weakness. All questions relating to experiences of discrimination in the survey used in this article are designed to limit chances of bias in response.

Previous research using self-reports in Ireland found higher rates of reported discrimination among national/ethnic minorities than among White Irish in 2004, in both looking for work and in the workplace – Black respondents reported particularly high levels of discrimination (O’Connell and McGinnity, 2008). This echoes findings by McGinnity et al. (2006) on the experience of racism and discrimination in a range of settings, including the workplace, where Black Africans reported the most discrimination of the immigrant groups. While we cannot rule out that there are ethnic/national differences in the propensity to report discrimination, previous research in the USA comparing self-reports of wage discrimination and actual wage penalties suggests that any ‘over-reporting’ was actually more likely among White respondents (Coleman et al., 2008). This suggests that ethnic/national differences, if anything, may be underestimated in this article.

Research questions

A major advantage of this article is that it is based on two national surveys that collected detailed information about the experience of discrimination, as well as a range of relevant socio-demographic indicators, during a booming economy and in the midst of a deep recession. Different theoretical perspectives generate different expectations about how the experience of discrimination might change. The data allow us to address a series of research questions comparing different types of discrimination experienced by Irish nationals and non-Irish nationals at different phases of the business cycle.

A key first question is whether non-Irish nationals experience higher rates of discrimination in the labour market than Irish nationals. In the light of the theoretical discussion and previous research on discrimination in Ireland, and internationally, our first hypothesis is that non-Irish nationals experience higher rates of discrimination than Irish nationals, both while looking for work and in the workplace.

Second, we expect to find variation in the extent of discrimination between groups. Approaches to discrimination highlighting racial prejudice would suggest that visibly different groups, Black Africans and Asians, and non-White Europeans experience greater discrimination. However, approaches that emphasize economic competition might suggest that NMS nationals would also experience discrimination, particularly while looking for work, as they constitute the largest group of immigrants in the labour market during the period. Furthermore this group tend to have lower levels of educational attainment than other immigrants, and may be competing for low-skilled jobs. While non-EU immigrants have the clearest restrictions on conditions of work, selective immigration policies, as noted above, mean that they tend to be more highly educated than NMS nationals, whose immigration is not restricted.

Our final set of questions relates to change over time. Here we encounter uncertainty because, between 2004 and 2010, the number and proportion of immigrants in the labour market increased substantially, and Ireland experienced a deep recession. Given the severity of the economic shock and the deterioration in the labour market, we might expect that discrimination against non-Irish nationals increased. With applications far exceeding vacancies, employers can ‘afford’ to select candidates on the basis of nationality/ethnicity. This would be consistent with in-group favouritism and economic competition approaches. It would also be consistent with a decline in openness to immigration and in willingness to accept immigrants of different race/ethnicity, and from poorer countries, observed in the Irish population between 2006 and 2010 (McGinnity et al., 2013). However, these negative tendencies may be offset to the extent that, over time, employers become more familiar with immigrants – as suggested by statistical discrimination approaches. Whether such familiarity is sufficient to counteract the impact of in-group favouritism and increased conflict over resources is unclear. Economic approaches highlight that discrimination entails costs for employers. Using ethnicity or nationality as a basis for recruiting or rewarding workers, rather than human capital differences, is an inefficient use of human resources – a luxury that employers may ill-afford in a recession.

Methodology

In this study we use self-reports of discrimination to measure the experience of discrimination while looking for work, and in the workplace in Ireland. We draw on two large-scale nationally representative surveys on the experiences of discrimination, carried out by Ireland’s Central Statistics Office (CSO) in 2004 and 2010. These surveys were collected as special modules of the Quarterly National Household Survey (QNHS), which is the official source of labour market indicators for Ireland. The Equality modules asked individuals whether they had experienced discrimination across a range of life domains over the previous two years. Respondents were shown the legal definition of discrimination in Ireland, and informed that when the term discrimination is used it refers to this legal definition only (CSO, 2011).⁴ Our analysis focuses specifically on two questions relating to self-reports of work-based discrimination:

In the past two years, have you personally felt discriminated against in the workplace?

Yes

Not applicable (don’t work, haven’t been working in the past two years)

Don’t know

In the past two years, have you personally felt discriminated against while looking for work?

Yes

Not applicable (don’t work, haven’t been looking for work in the past two years)

Don’t know

We restrict the analysis to the working age population (18–64). We base all analysis on the eligible population: we exclude respondents who answered ‘not applicable’ to the question. The Equality modules are particularly useful for our purposes because, unusually in Irish official statistics, they also collect information on ethnicity.⁵ Our analysis combines nationality and ethnicity to form national-ethnic groups.⁶ These groups are created on the grounds that they are comparable, have similar access to the Irish labour market and exhibit similar labour market trajectories.⁷ Merging ethnicity with nationality generates eight national-ethnic groups, outlined in Table 1:⁸ White Irish; White UK; White EU-13;⁹ White EU new member state (NMS); White non-EU; Black African; Asian; and minority ethnicity EU.¹⁰ The ‘White non-EU’ category refers to people of White ethnicity from a range of countries outside the EU, including North America, Australia and Asia. This group is predominantly English-speaking, and can be expected to share similar labour market experiences. The ‘minority ethnicity EU’ group consists of all non-White Europeans, including Black, Asian and ‘Other’ Irish nationals,¹¹ and the ‘Asian’ and ‘Black African’ groups are non-EU.¹²

Table 1.

National-ethnic groups, 2004 and 2010.

National-ethnic groups	2004		2010
	n	%	n	%
White:
Irish	23,047	93.6	15,095	89.7
UK	475	1.9	355	2.1
EU-13	188	0.8	147	0.9
EU NMS	161	0.7	644	3.8
Non-EU	275	1.1	150	0.9
Black African	97	0.4	111	0.7
Asian	100	0.4	104	0.6
Minority ethnicity EU	144	0.6	119	0.7
Subtotal	24,487	99.5	16,725	99.4
Unallocated residual	114	0.5	82	0.5
Ethnicity missing	9	0.0	14	0.1
Total	24,610	100.0	16,821	100.0

Source: QNHS Equality Module, 2004 and 2010.

Our primary objective is to evaluate the association between discrimination and ethnicity, nationality and recession. We examine whether, if given the same characteristics as the White Irish group, non-Irish nationals report labour market discrimination at higher rates than Irish nationals.

We use binary logistic regression models for our analysis, separate models are run for 2004 and 2010, and a pooled model of 2004 and 2010 data with interaction terms is used to test for significant differences over time. As the coefficients of standard logistic models are sensitive to bias due to inability to account for unobserved heterogeneity, as estimates are affected by omitted variables, we report the average marginal effects (AME). AME average the conditional effects, which means that they are invariant to the exclusion of covariates that are unrelated to covariates already in the model; they are also comparable across groups, samples, time and models (Mood, 2010). We base our test for statistical significance of the interaction effect on the estimated cross-partial derivative (Norton et al., 2004).¹³

Discrimination when looking for work and discrimination in the workplace are the dependent variables of our study. The binary variable is coded 1 if the person has experienced discrimination, 0 if not. All analysis is based on the population ‘at risk’. The population at risk of discrimination while looking for work is the working age population (aged 18–64) who were at work or looking for work in the past two years. The analysis of discrimination in the workplace is confined to those who were employees at the time of the survey.¹⁴

Our key focus is on differences in the experience of discrimination across national-ethnic groups. Our models control for gender, age, education and duration of residence in country as they are all considered in the literature to be potentially influential covariates.

We expect newly arrived immigrants to experience higher unemployment rates regardless of the business cycle (Wheatley Price, 2001). This disadvantage is expected to decline as immigrants gain more knowledge and experience of, and establish networks in, new labour markets (Brekke and Mastekaasa, 2008; Chiswick et al., 1997). However, previous Irish research has not found evidence of occupational assimilation as a function of time spent in the country (Barrett and Duffy, 2008). Recent migration is measured by including a dummy variable coded 1 if the individual has been resident in Ireland for two years or less, and 0 if they have been resident for longer.¹⁵

Aside from the control for duration, all control variables included are for the entire working population aged 18–64, so any effects of age, gender and so on are based on the full sample, and reflect patterns among the majority White Irish population. In the models of discrimination in the workplace we control for sector of employment, as working conditions may affect experience of discrimination.

A logistic regression model with interaction terms is run on pooled data in order to test for significant differences over time. In non-linear models the interaction effect is conditional on the independent variables, and, consequently, the significance of the coefficient is unreliable. Therefore we use the Stata user-written command ‘inteff’ to compute the correct marginal effect of a change in the two interacted variables, which calculates the statistical significance of the entire cross derivative (Norton et al., 2004). Significant interaction effects indicate that discrimination has increased or decreased for a national-ethnic group between 2004 and 2010.

By controlling for certain characteristics we can evaluate how work-based discrimination varies, and assess which groups are more vulnerable to discrimination. The models allow for us to investigate the effects of combinations of these characteristics, ensuring that some possible influences, net of discrimination, are controlled for. Crucially, they allow comparisons with Irish nationals. However, this is not a dedicated survey of immigrants, so it does not include all relevant variables in the analyses such as host language proficiency, ethnically constrained social networks and declining work motivation due to expectations of discrimination, all of which have been linked to immigrants’ experience in the labour market (Perreira et al., 2007). It is important to note that rates of discrimination reported in this module are based on the perception of the respondents, as discussed above.

Results

Table 2 shows rates of self-reported discrimination, when looking for work and in the workplace, in 2004 and 2010. Overall, just 6 per cent of respondents reported having experienced discrimination when looking for work in 2004 and 2010, and about 5 per cent reported having experienced discrimination in the workplace. Rates of discrimination were substantially higher among most non-Irish nationals in both periods, although discrimination fell for most non-Irish national groups between 2004 and 2010, particularly in looking for work.

Table 2.

Discrimination looking for work and in the workplace (%).

	Looking for work		In the workplace
	2004	2010	2004	2010
White Irish	5.0	5.3	4.3	4.5
Minority EU	9.4	17.0	16.3	11.2
White EU-13	10.6	5.2	9.0	4.6
White NMS	8.4	6.5	11.1	9.4
White non-EU	18.7	7.3	10.4	11.1
Asian	7.7	6.6	17.9	11.3
Black African	20.7	22.6	11.6	28.5
White UK	8.6	8.3	6.5	5.6
All	5.7	5.8	4.8	5.2

Source: QNHS Equality Module, 2004 and 2010.

Rates of discrimination when looking for work and in the workplace rose for Black Africans, and in 2010 Black Africans experienced extremely high rates of discrimination both looking for work (23%), and in the workplace (29%).

Regression analysis of discrimination when looking for work

Table 3 shows the results of a logistic regression model of discrimination when looking for work. The model controls for gender, age, education, unemployment, inactivity and duration of residence in Ireland. The results confirm that non-Irish nationals did experience significant rates of discrimination compared with the Irish group, in both 2004 and 2010. The decrease in the marginal effect for the non-Irish group indicates that discrimination had decreased in 2010; the interaction effect shows that this change over time is significant. This does not support our expectation that discrimination increased during the recession, although discrimination was still higher for the non-Irish group in 2010.

Table 3.

Logistic regression with average marginal effects (AME) – discrimination when looking for work.

	2004		2010		Δ 2004–2010
	Dy/dx	SE	Dy/dx	SE	Significantdifference
Ref: White Irish
Non-Irish	0.06***	0.01	0.03***	0.01	Yes
Ref: Male
Female	–0.02***	0.01	–0.00	0.01	No
Ref: Resident > 2 years
Resident < 2 years	–0.01	0.01	0.04*	0.02	No
Ref: 25–44
Under 25	0.00	0.01	–0.00	0.01	No
Age 45–64	0.00	0.01	0.03***	0.01	No
Ref: No formal/primary education
Lower secondary	–0.03***	0.01	0.02	0.01	No
Upper secondary	–0.04***	0.01	0.00	0.01	No
Post-secondary	–0.02**	0.01	0.01	0.01	No
Ref: Employed
Unemployed	0.11***	0.01	0.09***	0.01	Yes
Inactive	0.04***	0.01	0.06***	0.01	No
Constant	–2.82***	0.15	–3.99***	0.22	No
Pseudo R squared	0.10			0.07
N of cases	7334			5388

Notes: Dy/dx for factor levels is the discrete change from the base level; SE is the standard error. Significance probabilities for the coefficients: ***p < .001; **p < .01; *p < .05.

Source: QNHS Equality Module, 2004 and 2010.

We also find that while current unemployment had a strong positive association with the experience of discrimination while looking for work, this had decreased slightly in 2010 and this change over time is significant. The inactive group were also significantly more likely to experience discrimination when looking for work in 2004 and 2010. Females were less likely to experience discrimination in 2004 but not in 2010, the 45–64 age group were more likely in 2010. Those resident in Ireland for two years or less were more likely to experience discrimination in 2010, so recent entrants to a booming labour market did not report higher levels of discrimination than earlier immigrants, but recent arrivals during the recession did, perhaps reflecting the severe deterioration in the labour market. All education groups were significantly less likely to experience discrimination compared to the primary educated group in 2004.

Table 4 shows summary results of discrimination when looking for work for different national-ethnic groups. The model confirms that Black Africans encountered very high rates of discrimination in both 2004 and 2010. While there was some increase in discrimination experienced by Black Africans over time, this is not statistically significant. EU nationals of minority ethnicity also reported high levels of discrimination and this may have increased over time, although the increase is not statistically significant. These two groups vary in nationality but share minority ethnicity, suggesting that ethnicity is a common factor in their experience of discrimination.

Table 4.

Discrimination when looking for work among national-ethnic groups, summary.

	2004		2010		Δ 2004–2010
	Dy/dx	SE	Dy/dx	SE	Significantdifference
Ref: White Irish
Minority EU	0.05**	0.21	0.09***	0.02	No
White EU-13	0.06***	0.02	0.01	0.03	No
White NMS	0.04	0.02	0.01	0.01	No
White non-EU	0.08***	0.01	0.00	0.03	Yes
Asian	0.04	0.03	0.00	0.04	No
Black African	0.08***	0.02	0.10***	0.02	No
White UK	0.04***	0.01	0.01	0.02	No
Constant	–2.82***	0.15	–3.98***	0.23
Pseudo R squared	0.10		0.09
N of cases	7334		5388

Notes: Dy/dx for factor levels is the discrete change from the base level; SE is the standard error. Other covariates reported in Table 3 controlled for. Significance probabilities for the coefficients: ***p < .001; **p < .01; *p < .05.

Source: QNHS Equality Module, 2004 and 2010.

White non-EU nationals reported high rates of discrimination in 2004, but not in 2010, and this change is statistically significant. This is the only group that experienced a decline in their share of the population in the timeframe. We cannot rule out that those who were discriminated against left the country. The effect may also reflect a shift in the composition of non-EU immigrants in the Irish labour market following a policy change after EU enlargement. Non-EU immigrants working in Ireland in 2010 were more likely to have been recruited into high-skilled occupations with identified skills shortages, and thus encountered less discrimination than in 2004. However this effect does not hold for the Black African group, strengthening the argument that ethnicity is particularly salient when looking for work.

White EU-13 nationals and UK nationals experienced higher discrimination than Irish nationals in 2004, but this appears to have declined by 2010. While this change is not significant, it would be consistent with a statistical discrimination interpretation in which employers become increasingly familiar with European workers and their qualifications and skills.

Regression analysis of discrimination in the workplace

Table 5 shows that non-Irish nationals were more likely to experience discrimination in the workplace in 2004 and 2010 even after controlling for a range of factors. The average marginal effects for non-Irish nationals show that discrimination remained constant over time, and the change is not statistically significant. Again this does not support the expectation that discrimination among non-Irish nationals would increase during recession. We find that females experienced higher rates of discrimination at work throughout the period. We find a weak effect of sector, with a higher risk of experiencing discrimination in transport in 2004, and a lower risk in construction in 2004, and no significant sectoral effects in 2010. Sector does not influence the differences between national-ethnic groups in their experience of discrimination.¹⁶

Table 5.

Logistic regression with average marginal effects (AME), discrimination in the workplace.

	2004		2010		Δ 2004–2010
	Dy/dx	SE	Dy/dx	SE	Significantdifference
Ref: White Irish
Non-Irish	0.04***	0.01	0.04***	0.01	No
Ref: Male
Female	0.02***	0.01	0.03***	0.01	No
Ref: Resident > 2 years
Resident < 2 years		0.01	–0.02	0.02	No
Ref: 25-44
Under 25	–0.01	0.01	–0.01	0.01	No
Age 45–64	0.00	0.01	–0.00	0.01	No
Ref: No formal/primary education
Lower secondary	–0.02**	0.01	0.00	0.01	No
Upper secondary	–0.01	0.01	–0.02	0.01	No
Post-secondary	0.01	0.01	0.00	0.01	No
Ref: Industry
Agriculture/Forestry	0.00	0.02	–0.04	0.05	No
Construction	–0.04**	0.02	–0.03	0.03	No
Trade	–0.01	0.01	0.01	0.01	No
Transport	0.03**	0.01	0.02	0.02	No
Accommodation/Food	–0.00	0.01	0.01	0.01	No
Information/Communication	–0.01	0.02	0.01	0.02	No
Financial	0.01	0.01	0.01	0.02	No
Professional/Scientific	0.01	0.01	–0.01	0.02	No
Administration/Support	0.01	0.01	–0.00	0.02	No
Public administration/Defence	0.01	0.01	0.02	0.01	No
Education	0.01	0.01	–0.02	0.10	No
Health/Social work	–0.01	0.01	0.01	0.01	No
Other	–0.01	0.01	0.01	0.01	No
Constant	–3.15***	0.19	–3.37***	0.26
Pseudo R squared	0.02			0.03
N of cases	9987			6428

Notes: Dy/dx for factor levels is the discrete change from the base level; SE is the standard error. Significance probabilities for the coefficients: ***p < .001; **p < .01; *p < .05.

Source: QNHS Equality Module, 2004 and 2010.

Table 6 demonstrates that discrimination in the workplace is quite pervasive and persistent. The Minority EU, Asian, White NMS, Black African and White non-EU groups all experienced discrimination in the workplace in 2004. Discrimination persisted in 2010 for the White NMS, White non-EU and Black African groups. There is some indication that rates of discrimination fell for some groups, but the decline is not statistically significant, so contrary to our expectations, there is no evidence to suggest that discrimination in the workplace increased during the recession.

Table 6.

Discrimination in the workplace among national-ethnic groups, summary.

	2004		2010		Δ 2004–2010
	Dy/dx	SE	Dy/dx	SE	Significantdifference
Ref: White Irish
Minority EU	0.06***	0.02	0.04	0.03	No
White EU-13	0.03	0.02	0.01	0.03	No
White NMS	0.07***	0.02	0.04***	0.01	No
White non-EU	0.05**	0.02	0.05**	0.02	No
Asian	0.06**	0.02	0.04	0.02	No
Black African	0.08**	0.03	0.08***	0.02	No
White UK	0.02	0.02	0.01	0.02	No
Constant	–3.21***	0.19	–3.40***	0.27
Pseudo R squared	0.03		0.03
N of cases	9987		6428

Notes: Dy/dx for factor levels is the discrete change from the base level; SE is the standard error. Other covariates reported in Table 5 controlled for. Significance probabilities for the coefficients: ***p < .001, **p < .01, *p < .05.

Source: QNHS Equality Module, 2004 and 2010.

The Black African group showed much higher rates of discrimination than among White Irish in 2004 and 2010. It is clear from this study that the Black African group are faring particularly badly in the Irish labour market both when looking for work and in the workplace and there is no evidence to suggest that this is due to economic sector or occupation. Part of their manifest disadvantage may be attributed to the long-term effects of an asylum system that consigns asylum seekers to protracted periods of exclusion from Irish society and the labour market. This group of immigrants usually has less favourable labour market outcomes due to less positive selection processes, and greater difficulties in adapting to new environments resulting from stressful experiences surrounding their migration (Fleischmann and Dronkers, 2010). Unfortunately, the QNHS does not provide information on the visa/residency status of non-Irish nationals, so we cannot measure how many Black Africans are refugees, nor relate respondents’ experience of discrimination to residency status. It could also be, however, that employers assume that Black Africans were asylum seekers and had long periods out of the labour market, even if this is not the case – a form of stereotyping.

The White UK and EU-13 groups do not differ significantly from the White Irish in reported experience of discrimination. UK nationals have been coming to Ireland for decades and therefore may be less likely to experience discrimination in the workplace because they are more integrated. EU-13 nationals are a more recent but privileged group of immigrants in Ireland, typically highly skilled and with higher average incomes than Irish nationals (McGinnity et al., 2013).

We can reject the expectation that ethnic competition led to an increase in discrimination, as reports of discrimination in the workplace have remained relatively stable over time, or decreased for some groups.

Discussion

This article investigates the experience of discrimination in the Irish labour market. We examine the extent to which discrimination varies across different national-ethnic groups, and whether discrimination increases between 2004, during an economic boom, and 2010, in the midst of severe recession. To our knowledge, this is the first research to examine the impact of economic crisis on the experience of self-reported discrimination in the labour market.

We find that, overall, non-Irish nationals do experience higher rates of discrimination in looking for work, and in the workplace, in both boom and recession. We find substantial variation in discrimination across national-ethnic groups. In looking for work, ethnicity is particularly important, and we find that Black Africans and EU nationals of minority ethnicity are particularly likely to experience this form of discrimination. In the workplace, we find that most national-ethnic groups, apart from White UK and White EU-13 groups, are more likely than White Irish to experience discrimination in 2004. By 2010 the Black African, White NMS and White non-EU groups experience more discrimination than White Irish nationals. The finding that non-Irish nationals experience higher rates of discrimination than Irish nationals is consistent with previous research on immigrants’ experience of discrimination in Ireland (McGinnity et al., 2006; O’Connell and McGinnity, 2008).

Contrary to our expectations, we do not find that discrimination increased significantly in the context of recession and a growing immigrant population. In looking for work, the gap in reported discrimination between non-Irish nationals and White Irish fell between 2004 and 2010. In the workplace the gap between non-Irish nationals and White Irish remained relatively stable. We can thus reject the hypothesis that a labour market crisis and an increase in the proportion of immigrants lead to an increase in perceived ethnic competition for jobs and thus to an increase in discrimination against immigrants.

Why did reports of discrimination not increase during recession? Perhaps by 2010 immigrants have gained more experience and knowledge of the Irish labour market, and have established networks (Brekke and Mastekaasa, 2008). Attitudes to immigrants and immigration in Ireland had become more negative by 2010 (McGinnity et al., 2013), but immigrants were not, for the most part, scapegoated in public debates during the recession. It may also be that those who perceive ethnic competition are those who are unemployed, or inactive: this analysis concentrates on discrimination by employers and employees, thus ruling out discrimination by this group.

Why do reports of discrimination in recruitment among non-Irish nationals fall on average? In 2004 Ireland was a relatively new country of immigration, and employers may not have had experience with immigrant groups. By 2010 this would have changed. The fall in discrimination while looking for work provides some support for a statistical discrimination approach: over time as immigrant groups become more established, employers become better able to identify the work-related characteristics of immigrant job applicants. However, as we are not using panel data we cannot tell if any patterns of improved outcomes are the result of integration, cohort effects, selective out-migration, particularly of East Europeans, or of changing immigrant (self-) selection (Barrett and Duffy, 2008).

However, this is not true of all groups: visibly different ethnic groups, in particular Black African and minority ethnicity EU groups report very high rates of discrimination when looking for work, and their experience of discrimination did not decrease over time. This is consistent with discrimination based on racial prejudice, and a preference for White immigrants. Whatever the explanation, this finding, combined with high rates of unemployment and low rates of employment among these groups suggest they are particularly vulnerable.

Footnotes

Funding

This article represents a further development of research initially co-funded by the Irish Equality Authority and the European Union’s PROGRESS programme (2007–2013): the authors would like to acknowledge this funding. We would also like to acknowledge useful comments from participants in seminars at the ESRI, Dublin, and the University of Chapel Hill, North Carolina, and in paper sessions at the annual meetings of the European Consortium for Sociological Research (Tilburg, 2013) and the American Sociological Association (San Francisco, 2014).

Notes

Gillian Kingston is a Research Assistant at the Economic and Social Research Institute (ERSI) Dublin, and a PhD student, and teaching assistant in sociology at Trinity College Dublin. Most of her research to date has focused on migration, integration, labour market inequality, racial discrimination and gender.

Frances McGinnity is a Senior Research Officer and joint programme co-coordinator of Equality and Integration Research at the Economic and Social Research Institute; she is also Adjunct Associate Professor at Trinity College Dublin. Most of her research to date has examined labour market inequality – unemployment, temporary employment, part-time work, gender and racial discrimination. She is also interested in work–life balance, time-use and the integration of migrants in Ireland.

Philip J O’Connell is Director of the UCD Geary Institute for Public Policy and Professor of Applied Social Science at University College Dublin. Most of his work focuses on the labour market and on migration. He has an enduring interest in equality at work and in access to employment, publishing papers on wage inequality, on working conditions and workplace practices, on the transition from unemployment to work, on the determinants and effects of work-related education and training, and on the experiences of migrant workers.

References

Al Ramiah

Hewstone

Dovidio

Penner

(2010) The social psychology of discrimination: theory, measurement and consequences. In: Bond

McGinnity

Russell

(eds) Making Equality Count: Irish and International Research Measuring Equality and Discrimination. Dublin: Liffey Press, 84–112.

Barrett

Duffy

(2008) Are Ireland’s immigrants integrating into its labour market? International Migration Review 42(3): 597–615.

Barrett

McCarthy

(2007) The earnings of immigrants in Ireland: results from the 2005 EU survey of income and living conditions. IZA Discussion Paper no. 2990, August. Bonn: Institute for the Study of Labor (IZA).

Becker

(1957/1971) The Economics of Discrimination, 2nd Edition. Chicago, IL: University of Chicago Press.

Blank

Dabady

Citro

(eds) (2004) Measuring Racial Discrimination. Washington, DC: The National Academies Press.

Blau

(1977) Inequality and Heterogeneity: A Primitive Theory of Social Structure. New York: Free Press.

Blumer

(1958) Race prejudice as a sense of group position. Pacific Sociological Review 1(1): 3–7.

Brekke

Mastekaasa

(2008) Highly educated immigrants in the Norwegian labour market: permanent disadvantage? Work, Employment and Society 22(3): 507–26.

Central Statistics Office (CSO) (2011) Quarterly National Household Survey: Equality, Quarter 4 2010. Dublin: CSO.

10.

Chiswick

Cohen

Zach

(1997) The labor market status of immigrants: effects of the unemployment rate at arrival and duration of residence. Industrial and Labor Relations Review 50(2): 289–303.

11.

Coenders

Lubbers

Scheepers

Verkuyten

(2008) More than two decades of changing ethnic attitudes in the Netherlands. Journal of Social Issues 64(2): 269–85.

12.

Coleman

Darity

Sharpe

(2008) Are reports of discrimination valid? Considering the moral hazard effect. American Journal of Economics and Sociology 67(2): 149–75.

13.

Esses

Dovidio

Jackson

Armstrong

(2001) The immigration dilemma: the role of perceived group competition, ethnic prejudice, and national identity. Journal of Social Issues 57(3): 389–412.

14.

Fleischmann

Dronkers

(2010) Unemployment among immigrants in European labour markets: an analysis of origin and destination effects. Work, Employment and Society 24(2): 337–54.

15.

Hoynes

Miller

Schaller

(2012) Who suffers during recessions? Journal of Economic Perspectives 26(3): 27–48.

16.

Joyce

Quinn

(2014) The Organisation of Reception Facilities for Asylum Seekers in Ireland. Dublin: European Migration Network (EMN), ESRI.

17.

Khattab

Johnston

(2013) Ethnic and religious penalties in a changing British labour market from 2002 to 2010: the case of unemployment. Environment and Planning 45: 1358–71.

18.

McGinnity

Lunn

(2011) Measuring discrimination facing ethnic minority job applicants: an Irish experiment. Work, Employment and Society 25(4): 693–708.

19.

McGinnity

O’Connell

Quinn

Williams

(2006) Migrants’ Experience of Racism and Discrimination in Ireland. Dublin: ESRI.

20.

McGinnity

Quinn

Kingston

O’Connell

(2013) Annual Monitoring Report on Integration 2012. Dublin: The Integration Centre and the ESRI.

21.

Mood

(2010) Logistic regression: why we cannot do what we think we can do, and what we can do about it. European Sociological Review 26(1): 67–82.

22.

Norton

Wang

(2004) Computing interaction effects and standard errors in logit and probit models. STATA Journal 4(2): 154–67.

23.

O’Connell

(2013) Cautious adjustment in a context of economic collapse: the public sector in the Irish crisis. In: Vaughan-Whitehead

(ed.) Public Sector Shock: The Impact of Policy Retrenchment in Europe. Cheltenham: Edward Elgar, 337–70.

24.

O’Connell

Joyce

(2013) International migration in Ireland, 2012. Geary Working Paper 2013/04. Dublin: UCD Geary Institute.

25.

O’Connell

McGinnity

(2008) Immigrants at Work: Ethnicity and Nationality in the Irish Labour Market. Dublin: Equality Authority.

26.

Organisation for Economic Co-operation and Development (OECD) (2013) International Migration Outlook 2013. Paris: OECD Publishing.

27.

Pager

Shepherd

(2008) The sociology of discrimination: racial discrimination in employment, housing, credit, and consumer markets. Annual Review of Sociology 34: 181–209.

28.

Perreira

Harris

Lee

(2007) Immigrant youth in the labor market. Work and Occupations 34(1): 5–34.

29.

Phelps

(1972) The statistical theory of racism and sexism. American Economic Review 62(4): 659–61.

30.

Quillian

(2006) New approaches to understanding racial prejudice and discrimination. Annual Review of Sociology 32: 299–328.

31.

Rafferty

(2012) Ethnic penalties in graduate level over-education, unemployment and wages: evidence from Britain. Work, Employment and Society 26(6): 987–1006.

32.

Reskin

(2000) The proximate causes of employment discrimination. Contemporary Sociology 29(2): 319–28.

33.

Reskin

(2003) Including mechanisms in our models of ascriptive inequality: 2002 presidential address. American Sociological Review 68: 1–61.

34.

Russell

McGinnity

(2011) Workplace Equality in the Recession? The Incidence and Impact of Equality Policies and Flexible Working. Dublin: ESRI.

35.

Schneider

(2008) Anti-immigrant attitudes in Europe: outgroup size and perceived ethnic threat. European Sociological Review 24(1): 53–67.

36.

Wheatley Price

(2001) The employment adjustment of male immigrants in England. Journal of Population Economics 14(1): 193–220.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.19 MB