Mass immigration into Ireland is a new phenomenon.
- As recently as 1991, Irish residents born elsewhere numbered 228,725 – or 6% of the total population – and of those, only 40,341 had been born outside the UK or the US.
- Two decades later, the foreign-born numbered 766,770 – or one-sixth of the total – and three-fifths of those (or one-tenth of the total) came from outside the UK.
The huge rise in the numbers of residents of east European origin is often highlighted (Ruist 2013), but between 2002 and 2011 the number of African-born residents doubled (to 54,419) and that of Asian-born residents almost trebled (to 79,021). The influx in the 2000s was also, in relative terms, extremely high by west European standards (Figure 1).
Figure 1 Immigration rates in western Europe, 2002−2011
At first sight the impact of immigration on Irish public opinion is curious and ambivalent.
- On the one hand, at least so far, Ireland has been spared the xenophobic brand of politics currently in the ascendant across much of Europe – Ireland lacks a Front National or a Lega Nord. Moreover, successive Eurobarometer polls since the early 2000s suggest that the Irish worry much less about immigration than citizens of neighbouring European economies.
- On the other hand, opinion polls also point to significant anti-immigrant feeling in Ireland. The Irish National Election Study, a panel survey conducted between 2002 and 2007, indicated that hostility to immigrants was strongly correlated with hostility to travellers, implying that apart from any economic threat they presented, immigrants were perceived by some as undesirable, as ‘others’ or ‘different’. The Irish National Election Study also indicated that education was positively correlated with pro-immigrant attitudes (Denny and Ó Gráda 2013).
The European Social Survey (ESS) is a population representative cross-national survey that has been conducted every two years across the continent since 2002. Over 30 countries currently participate. Typically, data collection takes place over a period of about eight months, spanning two calendar years.
The ESS contains six questions about immigrants: three about how many immigrants should be allowed in (depending on race, country of origin, etc.), and three more general questions about whether the respondents thought immigration were good for the country in different domains. Using factor analysis, we used these six questions to generate a synthetic measure of whether people were for, or against immigrants, and immigration in general. Normalised to a mean of zero and a variance of 1 over the three waves, our measure can be used to analyse the trend in Irish attitudes to immigrants, and what sort of people are more, or less sympathetic to immigrants.
We generated a second variable, Z, to measure xenophobia, and it attempts to capture that particular hostility reserved for immigrants who differ ethnically/racially from the host population. The ESS first asks respondents about their attitude to immigrants from the same race/ethnicity as the majority of residents. It then asks the same question about immigration from different race/ethnic groups than the majority. The possible responses to both questions were “Allow many to come and live here”, “Allow some”, “Allow a few”, and “Allow none”. We set xenophobia to one (Z=1) if respondents want to allow fewer from the non-majority race/ethnic group than from the majority, and equal to zero otherwise.
In the Irish context, which is our sole focus here, Z may be interpreted as a measure of a respondent’s preference for returning Irish immigrants, and for immigrants from the UK and the US, over immigrants from elsewhere. In the case of other economies, where return migration is unimportant, it might indicate instead a preference for Caucasian over black or Muslim immigrants – or, in the case of Israel, for Jewish over all other immigrants.
The ESS has already been widely used for insights into popular attitudes to immigration. Here we report the results of comparing the Irish response to immigration over the six ESS rounds so far (biannually from 2002 to 2012). The period coincides with the big rise in Irish immigration and also with the last years of the Celtic Tiger (2002-07) and its demise.
Figures 2a and 2b describe the shifts in our measure of public acceptability of migration (ATTIM) and xenophobia (Z) between 2002 (Round 1) and 2012 (Round 6). Between 2002 and 2006, as immigration rose rapidly, ATTIM rose in tandem. The sharp fall in the wake of economic collapse – Irish GDP fell by 13% between 2007 and 2010, and the unemployment rate rose from 4.8 to 13.9 – is perhaps not so surprising, but the reversal to 2012 is.
This poses two related questions.
- First, can the ESS tell us what factors influenced Irish attitudes to immigration between 2002 and 2012?
- Second, what caused those attitudes to change over time?
We describe the technical details of our answers to those questions elsewhere (Denny and Ó Gráda 2013). Here we provide a non-technical summary. Our analysis focuses on wave 1 (2002-03, just as mass immigration was beginning), wave 2 (2006-07, just before the economic bubble burst), and wave 3 (2012-13, when the Irish economy was still in deep crisis). For convenience, we refer to the data by the first year of each pair.
Some consistent patterns emerge. Women and rural dwellers were, and are, more hostile to immigration; the educated and the foreign-born less so. ATTIM rose with positive perceptions on the state of the economy, and fell with worries about household income. Throughout the decade those who held liberal views on gay and lesbian rights were more pro-immigrant. We included the ESS variable on this issue as a measure of a broader non-economic hostility to ‘others’. We also found that while the coefficients on some variables (e.g. years of education) remained quite steady over the entire period, others seemed quite sensitive to economic conditions. The latter included variables reflecting perceptions about the state of the economy and about income.
Z for xenophobia
Figure 2b describes the trend in Z, our measure of xenophobia, between 2002 and 2012. Note in particular the sharp fall between 2004 and 2006, when immigration was at its highest. The decline since – slow, but persistent – offers an interesting and rather surprising perspective on the Irish response to globalization in an era of recession. We also analysed the impact on Z of the same variables as ESS waves 1, 3, and 6. None of our variables packs much punch in either 2002 or 2006, but in 2012-13 the negative coefficient on STFECO suggests that the greater satisfaction with the state of the economy, the lower Z. Less easy to understand is the finding that the greater the difficulty in making ends meet, the higher Z. Again we find that in times of recession, those who tend to be hostile to lesbian and gay rights also tend to be hostile to immigrants of a different ethnicity.
Decomposing changes in attitudes
Irish attitudes to immigration hardened with the economic downturn, but not in a straightforward way. To what extent is that hardening explained by changes in economic wellbeing, and to what extent by a shift in preferences or – in the jargon of Gary Becker – in an increasing ‘taste’ for being anti-immigration?
In economics, the most straightforward way to answer this question is the decomposition procedure first employed four decades ago by Ronald Oaxaca in seeking to identify the impact of discrimination in the US labour market. This procedure allows us to measure the relative impact of changes in preferences, and changes in underlying economic conditions on ATTIM, both in 2002-06, and in 2006-12.
The technicalities are explained in Denny and Ó Gráda (2013). In brief, the decomposition into the endowment and ‘taste’ parameters suggest that in the earlier period both were about equally responsible for the change in ATTIM, while in the later period both components and their interaction had a significant role to play. Between 2006 and 2012, people’s falling satisfaction with the economy, as well as people’s own more difficult economic circumstances, explains most of the hardening of attitudes towards immigrants. Rising education levels did something to stem the tide. Interestingly, too, that there has been a changing effect of age – with a higher age switching from being a sympathetic factor towards immigrants in 2006, to having the opposite in 2012.
Over the past decade or so, Ireland has been transformed from a place where immigrants were few, to a place where one resident out of six is born outside the country. The impact of this change on public opinion is of considerable interest. Here, we have sought to identify that impact and the factors that influence it. Not surprisingly, the economic downturn after 2007 had a negative impact on attitudes to immigration. At the same time, there is evidence in the ESS data that the Irish have become more accepting of people from very different backgrounds. How the trends in Irish opinion have diverged from those of other European countries is an interesting question, which we hope to address in future work.
Card, D, C Dustman and I Preston (2012), “Immigration, wages, and compositional amenities”, Journal of the European Economic Association 10(1), pp. 78-119.
Denny, K and C Ó Gráda (2013), “Irish attitudes to immigration during and after the boom”, University College Dublin Centre for Economic Research Working Paper No. 13/18.
Oaxaca, R (1973), “Male-female wage differentials in urban labour markets”, International Economic Review 14, pp. 693-709.
O'Rourke, K H and R Sinnott (2006), “The determinants of individual attitudes towards immigration”, European Journal of Political Economy 22(4), pp. 838-61.
Ruist, J (2013), “Eastern European migrants are net contributors – not costs – in the West”, VoxEU.org, 17 September.