Massive Migration and Elections: Evidence from the Refugee Crisis in Greece
| Author | Chrysovalantis Vasilakis |
| Date | 01 June 2018 |
| Published date | 01 June 2018 |
| DOI | http://doi.org/10.1111/imig.12409 |
Massive Migration and Elections: Evidence
from the Refugee Crisis in Greece
Chrysovalantis Vasilakis*,**
ABSTRACT
This article explores whether the massive arrival of refugees at Greek islands has had an
impact on natives’voting behaviour. Our results show a positive and significant effect of refu-
gees’presence on votes for the Greek extreme-right party Golden Dawn. More precisely, we
find that a 1 per cent increase in the share of refugees is associated with an increase of 5 per
cent in the share of votes for Golden Dawn. This outcome is robust under different estimation
methodologies and placebo regressions.
INTRODUCTION
The political instability of the Middle East has prompted an unprecedented surge of refugees to
Europe. According to a report of the United Nations High Commissioner for Refugees, released in
June 2015, the number of forcibly displaced people around the globe had reached a figure of 59.5
million (UNHCR, 2015), constituting the largest number of refugees since WWII. Whereas refu-
gees have historically sought refuge in developing countries because of geographical proximity, the
recent turmoil in North Africa and the Middle East, alongside the continuing violence in Afghani-
stan, has sparked a massive increase in displacements toward the European Union, with the number
of people applying for asylum reaching a record figure of 1.26 million in 2015 (Eurostat, 2016).
Meanwhile, we have witnessed a rise in nationalist and anti-immigration parties in several Euro-
pean countries, including Austria (FP
€
O), Belgium (Vlaams Belang), Finland (True Finns), France
(Front National), Germany (Pegida and AfD), Sweden (Sweden Democrats), the United Kingdom
(UKIP), and the country studied in this article, Greece (Golden Dawn). While the rise in extreme-
right parties is not a new phenomenon in Europe (Arzheimer, 2009), the trend does seem to have
accelerated in more recent years, as epitomized by the successful Brexit campaign. This, in turn,
raises two related questions: (i) whether a massive refugee presence can affect the political choices
of natives in European countries, and (ii) whether increases in right-wing votes reflect anti-immi-
gration feelings going beyond purely economic concerns.
To address these questions, we explore the effect of the unprecedented surge of refugees to the
European Union on two parliamentary electoral results in Greece in January and September 2015.
Our focus on refugees in Greece contributes to the existing literature by allowing us to instrument
the number of refugees with the geographical distance separating (recipient) Greek islands from
(sending) Turkish coasts. We are thus able to clearly identify the positive effect of a massive
increase in refugees on extreme-right votes. Moreover, our analysis reveals that this same surge of
* University of Bangor
** IRES, Universite Catholique de Louvain
doi: 10.1111/imig.12409
©2017 The Author
International Migration ©2017 IOM
International Migration Vol. 56 (3) 2018
ISS N 00 20- 7985 Published by John Wiley & Sons Ltd.
refugees has negatively affected right-wing parties, thus suggesting a transfer of right-wing voters
toward more radical political platforms. Lastly, the high number of refugees, together with the short
timespan separating the two electoral rounds, allows us to conclude that the aforementioned effects
are likely to have been driven by a potential rise in anti-immigration feelings or a protest for feel-
ing abandoned by the Greek government of the Greek voters. To further rule out economic chan-
nels as the cause of the rise in political extremism, we conduct a placebo analysis. We substitute
the voting share of Golden Dawn in the 2015 election with that of 2012; given that this election
took place before the large inflow of refugees, we should see no effect of the presence of refugees
in 2012 on this earlier election. Indeed, we find no effect.
Immigration —especially in the presence of large numbers of migrants —has been shown to
generate anti-immigration feelings (Hopkins, 2010; Dustmann et al., 2011) and these feelings are
expressed through voting for extreme-right parties or racist behaviour (e.g., Knigge, 1998; Norris,
2005; Rydgren, 2005; Halla et al., 2017). The channels explaining this phenomenon are multiple.
Anti-immigration attitudes are explained by both economic factors, such as labour market competi-
tion and the impact of immigration on the fiscal burden, and non-economic factors rooted in cul-
tural values and beliefs (see, for example, Rydgren, 2005; Arzheimer, 2009; Hainmueller and
Hopkins, 2014; Barone et al., 2016; Halla et al., 2017).
Theories tying immigration to the labour market rely mostly on the Heckscher–Ohlin and the fac-
tor proportions models and emphasize the competition that native workers may be facing from
migrants. The main conclusions are that under fairly general conditions the skilled (un-skilled)
natives should be in favour of unskilled (skilled) migrants, while opposing skilled (unskilled) ones
(Scheve and Slaughter, 2001; Mayda, 2006; O’Rourke and Sinnott, 2006). The empirical support
shows that low-skilled immigration increases (decreases) negative attitudes toward migrants from
low-skilled (high-skilled) natives (Scheve and Slaughter, 2001; Mayda, 2006; O’Rourke and Sin-
nott, 2006; Facchini and Mayda, 2009). The effect of migrants on labour markets goes beyond
wage considerations, though, since higher unemployment rates have equally been shown to boost
extreme-right parties (Jackman and Volpert, 1996) or right-wing criminal activities (Falk et al.,
2011). In the specific Greek context, Karydis (2004) presents survey evidence according to which
anti-immigration feelings are rooted in the fear of migrants stealing jobs. On the other hand, there
is evidence in Germany that the migrants do not displace the jobs of native workers but rather
struggle to find jobs (Gehrsitz et al., 2017). The second economic channel —which has been
shown to be quantitatively bigger than that of the labour market (Dustmann and Preston, 2007) —
operates via the increased fiscal burden required to provide social security to the newcomers. Nev-
ertheless, there is evidence that economic conditions are not the main reason for the support of
extreme-right parties (see Labrianou and Ellinas, 2016; Rydgren, 2005; Ivarsflaten, 2005).
Setting aside pure economic factors, natives’attitudes toward economic migrants and refugees
have equally been shown to be affected by anti-immigration feelings and perceived threats to one’s
cultural identity (Mayda, 2006; Arzheimer, 2009). In the context of Sweden, for instance, Bauer
et al. (2000) conclude that this dimension is more salient than economic factors, while Dustmann
and Preston (2007) find that in Great Britain the cultural aspect is highly relevant in explaining
negative attitudes toward migrants. In fact, cultural distance has been shown to matter (Mayda
et al., 2015). According to Mendez and Cutillas (2014), more culturally proximate migrants (e.g.,
Latin Americans in Spain) may translate into more votes for leftist parties, while culturally distant
ones (e.g., Africans in Spain) increase votes for anti-immigration parties. Moreover, an experimen-
tal study of the Netherlands further confirms the salience of cultural distance being perceived as a
threat to cultural identity (Sniderman et al., 2004).
Two studies on the effect of immigration on Danish politics partially concur with a positive asso-
ciation with extreme-right-wing votes. Harmon (2014) uses historical renting housing stock to
instrument the current settlements of migrants and finds that higher ethnic diversity in Danish
municipalities increases the support for anti-immigration (nationalist) parties at municipal elections,
Massive Migration and Elections 29
©2017 The Author. International Migration ©2017 IOM
To continue reading
Request your trial