The benchmark analysis results, along with the effect of the moderator, and economic development on the relationship between refugee influx and voting behavior, are discussed in this section.
5.2.1 The effect of refugee influx on voting behavior
The aim of this study is to test whether short-term shock and refugee influx influence individuals' voting behavior. Therefore, a quasi-experimental study is conducted, using DID regression analysis. Treatment groups, districts above 1% refugee share in population is compared with districts below 1% refugee share in population. Table 4 reports the influence of the 2015 refugee influx on voting behavior. Each column shows a separate regression and presents corresponding estimates. For instance, column (1) shows the results for individuals who have voted for the Christian Democratic Party (CDU), column (2) Free Democratic Party (FDP), column (3) Social Democratic Party of Germany (SPD), and column (4) Alternative for Germany (AfD). The DID estimates illustrate the net effect of the refugee influx, which assimilates the channelling result of economic development, which we address in later sections.
Table 4
The effect of refugee influx on voting behavior
| Dependent variable: Voted in the German federal election |
| Incumbent party CDU (1) | FDP (2) | SPD (3) | Right wing AfD (4) |
Refugee share | -4.88*** (0.64) | -2.81** (0.43) | 3.76** (0.68) | 2.22 (0.33) |
Post influx | -0.02*** (0.01) | -0.01 (0.001) | -0.03*** (0.01) | 0.08*** (0.01) |
Refugee share * Post influx | 2.42*** (0.56) | 1.58*** (0.40) | -1.97*** (0.58) | -2.88*** (0.32) |
Gender (Female) | 0.02*** (0.004) | -0.01*** (0.000.1) | -0.01** (0.004) | -0.02*** (0.001) |
Age (years) | 0.003*** (0.00001) | 0.001*** (0.0001) | 0.002*** (0.0001) | -0.0002*** (0.00002) |
Partner (yes) | 0.05*** (0.01) | 0.01*** (0.002) | 0.01*** (0.01) | 0.01*** (0.002) |
State fixed effects | Yes | Yes | Yes | Yes |
Observations | 39,646 | 39,646 | 39,646 | 39,646 |
R-squared | 0.07 | 0.27 | 0.03 | 0.03 |
Note. This table shows the results of the difference-in-differences analysis. The observations are at the individual-level unit. The dependent variable measures whether the individual voted for CDU (column 1), FDP (column 2), SPD (column 3), AfD (column 4). Measured using SOEP question, which political party you have voted for in the previous federal election. Dummy variables are created of individuals who voted for (1) CDU (2) FDP (3) SPD (4) AfD. In parentheses are the robust standard errors. ∗Significant at 10% level; ∗∗significant at 5% level; ∗∗∗significant at 1% level.
From the results, it can be observed that the refugee influx in Germany has significantly influenced voting behavior in the federal election. The significant positive coefficients of 2.42 and 1.58 for the interaction terms for the incumbent party, CDU, and FDP, indicate an increase in the popularity of more liberal parties. For instance, individuals in districts with a 1% great refugee share are 2.42 times more likely to vote for CDU. Contrarily, negative significant coefficients of -1.97 and − 2.88 are observed for SPD and the right-wing party, AfD, showing a decrease in voting popularity for both the anti-immigration parties.
In general, voters in democracies have a pattern of either rewarding or punishing political parties in elections depending on their satisfaction with their party's decisions. To understand the influence of refugee influx on voting behavior, the different stances of these political parties on immigration and refugee influx need to be acknowledged (Ward, 2020). The incumbent party, Christian Democratic Party (CDU), opposed limiting residential permits for immigrants and refugees in Germany. The German Chancellor and leader of the CDU party, Angela Merkel, implemented an open-door policy in 2015, allowing millions of Syrian refugees to enter and reside in Germany. The party allowed family members of accepted asylum seekers permission to follow them and live together in Germany. CDU worked towards helping asylum applicants find jobs prior to their visa, to cover their living expenses in Germany (Abdou et al., 2022; Hertner, 2022). Though FDP was more critical of the open-door policy by Angela Merkel, and emphasized more efficient management of the refugee crisis, the party pushed for faster asylum procedures, refugee integration, and cooperation among European nations to address the refugee crisis (Bauder and Semmelroggen. 2009; Göpffarth, 2017). FDP proposed a point-based immigration system and a simpler immigration procedure to promote skilled immigration to Germany. Lastly, FDP addressed the need to allocate aid to the home nations of the refugees to fight the root cause of the refugee crisis. Both parties were more inclined and open toward foreigners and refugees entering Germany.
In contrarily, Social Democratic Party (SPD) party leaders of the SPD, were stricter, and raised concerns regarding Angela Merkel’s open-door refugee policy. They expressed apprehensions regarding security threat of accepting refugees, and suggested better security measures to refrain allowing ‘terrorists or criminals’ along with the refugees. The party leaders claimed that the refugee influx would create economic and social problems, such as the financial burden due to the social services and the employment shortage in the labor market. Lastly, they wanted a faster deportation system for rejected asylum applicants (Schmidtke, 2016; Boswell and Hampshire, 2017). The extreme right, the party with the strictest immigration law proposal, AfD, proposes a "zero immigration policy," implementing a cap on all immigration, creating net immigration of zero or negative. They have continuously called for the deportation of rejected asylum seekers and reforming the German constitution to ensure fewer people are granted asylum. The anti-immigration and Islamophobic party believe migrants or refugees from Muslim-majority countries threaten Germany's security and stability (Art, 2018; Arzheimer, K., & Berning, 2019).
The support for immigrant-friendly and disapproval of anti-immigration parties with the refugee influx indicate improvement in tolerance and acceptance of individuals towards refugees and immigrants. Theoretically, Gordon Allport's (1955) hypothesis that social contact between social groups is ample for reducing intergroup prejudice aligns with the findings of our study. Individuals are now more prone to interact with refugees or immigrants after the influx. In addition, the extensive empirical and experimental research also weighed how interaction challenges prejudice or intolerance by reducing intergroup anxiety and increasing empathy for other groups. For instance, prejudice or intolerance is developed from false beliefs, misconceptions, and stereotypes; therefore, interaction with outer groups reduces prejudice as false beliefs or misconceptions are erased. The consistency of the intergroup contact theory was robust to Pettigrew and Tropp's (2006) meta-analytic test. They tested 713 independent samples, among which 515 studies found an inverse relationship between contact and prejudice. These findings of improved support for immigrant-friendly parties, and disapproval of anti-immigrant parties are consistent with my previous chapter on tolerance, where the refugee influx made individuals more tolerant towards outer groups. These results are antithetical to the national election results from the 2013 to 2017 federal election, the exact timeline used in this paper, introducing an important finding of our paper. The DID regression analysis interaction coefficient explains how districts with refugee influx (treatment region) behave compared to districts with a minimal or negligible number of refugees. Therefore, contradictory results of acceptance and popularity of pro-immigration parties and disapproval of anti-immigration parties indicate that individuals who are more familiar with and likely to interact with refugees turn out to be more accepting or welcoming towards refugees. The change in national voting results or patterns might be due to individuals in control districts with lower refugee numbers who are less likely to interact with refugees or immigrants. Therefore, based on Stephan and Stephan's (2000) social threat theory, the probable reason for intolerance or unwelcoming attitude towards refugees and immigration in areas with minimum interaction with refugees can be understood to be formed from physical and social threats associated with outer groups (i.e., refugees). As discussed in section 3.1, the dangers can vary from cultural dilution to economic threats of losing jobs or competition in the labor force.
5.2.2 The effect of refugee influx on voting preference
In addition, whether the refugee influx has altered the political party preference of individuals is discussed. Table 5 shows the DID analysis results showing the refugee influx's effect on the voting preferences of individuals in Germany. The voting behavior and preference can be understood to be correlated or as equivalent measures for voting behavior.
Table 5
The effect of refugee influx on political party preference
| Dependent variable: Inclined towards political party |
| Incumbent party CDU (1) | FDP (2) | SPD (3) | Right wing AfD (4) |
Refugee numbers | -2.31 *** (0.42) | -1.60*** (0.29) | 2.37*** (0.43) | 0.48 (0.16) |
Post influx | -0.02*** (0.002) | -0.004 (0.003) | -0.03*** (0.004) | 0.03*** (0.002) |
Refugee numbers * Post influx | 1.51*** (0.37) | 0.94*** (0.27) | -1.56*** (0.37) | -0.84*** (0.17) |
Gender (Female) | -0.02*** (0.003) | -0.01*** (0.001) | -0.02*** (0.003) | -0.01*** (0.001) |
Age (years) | 0.003*** (0.00001) | 0.001*** (0.00004) | 0.003*** (0.0001) | 0.00004 (0.00003) |
Partner (yes) | 0.02*** (0.003) | 0.01*** (0.0001) | 0.001 (0.003) | 0.0003** (0.001) |
State fixed effects | Yes | Yes | Yes | Yes |
Observations | 53,522 | 53,522 | 53,522 | 53,522 |
R-squared | 0.05 | 0.13 | 0.04 | 0.01 |
Note. This table shows results of the difference-in-differences analysis. The observations are at the individual-level unit. The dependent variable measures whether the individual in inclined towards CDU (column 1), FDP (column 2), SPD (column 3), AfD (column 4). Measured using SOEP question, which political party are you inclined towards. Dummy variables are created of individuals who voted for (1) CDU (2) FDP (3) SPD (4) AfD. In parentheses are the robust standard errors. ∗Significant at 10% level; ∗∗significant at 5% level; ∗∗∗significant at 1% level.
Individuals likely vote for the political party they prefer; therefore, the results of this DID analysis using political party preference as the dependent variable is expected to show a similarly significant relationship with the refugee influx. The results show similar significant coefficients of voting preference with refugee influx for political parties such as CDU (2.42 and 1.51), FDP (1.58 and 0.94), SPD (-1.97 and − 1.56), and AfD (-2.88 and − 0.84). The voting behavior or preference results showed that voters changed their voting behavior with short-term shocks (i.e., refugee influx). This finding opposes the elite cue theory of party positioning, which claims that individuals tend to follow elite individuals, such as party leaders’ opinions regarding a particular issue (i.e., refugee influx or the refugee crisis). Instead, the voters let their policy preferences on issues shape their decision to support or vote for a specific political or governing party. The literature claims that voters tend to follow their policy preferences rather than elite cues when they assign high importance to the issue (Chzhen et al., 2014; Weber and Saris, 2015), showcasing how immigration and the refugee crisis are essential to the Germans.