The huge inflow of asylum seekers to European countries in the early 1990s drove those countries to initiate policies that restricted asylum seekers’ rights and benefits. Were these policies spontaneous responses to the mounting inflow or instead political outcomes determined by political factors such as partisanship and election timing? Analyzing data on the introduction of policies that restricted asylum seekers’ welfare benefits and rights in 13 European countries from 1981 to 2000, this article finds that upcoming elections increase the likelihood of policy introduction. This election effect is greater in the presence of right-wing parties. These results imply that policies regarding asylum seekers’ welfare benefits have been utilized by right-wing parties for electoral purposes. Therefore, these policies should be understood as the outcome of political choices.
As I said, the Bill was politically aimed at the pre-election period. The newspapers that give support and succour to the Tory party played their role well, not only by softening public opinion in advance of the House's consideration of the Bill, but by poisoning workers’ minds with their repeated stories of con men, of racketeering and of massive social security fraud.Footnote 1
A British Labour Party Member of Parliament, Dave Nellist, criticized the Conservative government when it introduced a bill in 1991 that was meant to restrict asylum seekers’ rights and argued the government was making use of the issue for electoral purposes. Although this quote illustrates only a piece of the burning debate on the asylum issue and the bill between the Labour Party and the Conservative Party in the British Parliament, it gives a challenge to a conventional understanding of restrictive policies on asylum seekers: the policies have been spontaneous responses to the increasing inflow of asylum seekers (Boswell, 2003; Crisp, 2003; Geddes, 2003; Meyers, 2004).
As the number of asylum seekers grew,Footnote 2 European countries invented or modified asylum policies to make asylum procedures stricter, discourage asylum-seeking people from coming to their countries, and eventually decrease the inflow of asylum seekers. For instance, they began to limit the eligibility requirements to apply for political asylum by adopting clauses such as ‘safe third country’, ‘manifestly unfounded asylum claim’ and ‘safe countries of origin’. They also made it harder and riskier for asylum seekers to come to host countries by imposing penalties on carriers who transported individuals lacking the travel documents. In addition, they believed that many asylum seekers were not genuine. They believed the asylum seekers left their home countries, not due to real fears of political or religious persecution, but in search of economic opportunities such as employment and welfare benefits. Thus, these countries created policies that withdrew or reduced asylum seekers’ welfare benefits and rights in order to discourage, ‘bogus’ asylum seekers.
These policies have been understood primarily as responses to the rising level of asylum seekers. Although bad economic conditions were sometimes considered as triggering the introduction of asylum-restricting policies as well (Boswell, 2003), some literature has suggested that ‘sheer numbers have played a part’ in policy modification (Crisp, 2003, p. 77). Political factors, such as party politics and electoral competition, were either not taken into account or were found to make no difference in asylum policies (Bloch, 2000).
However, taking a look at asylum issues in Europe encourages one to believe the introduction of restrictive policies was not simply spontaneous responses to the growing number of asylum seekers. Political actors have presented dissimilar stances on the asylum issue. Voters were divided on the issue along party lines: voters for center-right parties took more restrictive positions than those for center-left parties.Footnote 3 Political parties repeatedly failed to find consensus on asylum policies, either (Schuster, 2003). Consequently, they fiercely debated the issue in parliaments and criticized each other's positions in election campaigns (Steiner, 2000). The asylum issue instilled political conflicts between different regional units as well. Because the fiscal burden to accommodate asylum seekers often fell upon local governments, tensions arose between central and local governments, as well as between different local governments (Amrute and Pfohman, 2001).
Political asylum has created disputes between political parties, voters and regions in European countries because of the different stances political actors have taken on the issue. Given this, asylum policies should be considered as political outcomes that develop from political factors such as partisanship, electoral competition, and so on. Thus, this article studies political factors for the introduction of restrictive asylum policies in 13 European countries from 1981 to 2000 that have rarely been studied.Footnote 4 Regarding asylum policies, this article focuses on polices that withdrew and cut asylum seekers’ welfare benefits and rights. Also, as regards political factors, this article examines how government partisanship and election timing helped decide the likelihood of policy introduction.
Asylum policies in Europe have been ‘Europeanized’ since the beginning of the new millennium. As a result, the study of the development of domestic asylum policies may not be worthy any more. However, the role of political parties is a major determinant of European Union politics and policies, too (Hix, 2008). Thus, examining the position and function of political parties in formulating asylum policies on the country level will also be helpful for further research on the formation of asylum policies at the European Union level.
Literature Review on Asylum Policies
The literature on the formulation of asylum policies has considered the introduction of restrictive policies on asylum seekers as impulsive responses to increasing inflows of asylum seekers. Although some literature incorporated economic factors, such as bad economic conditions and fiscal burdens to accommodate asylum seekers (Bosswell, 2003; Geddes, 2003), most of these studies have not hesitated to emphasize the role increasing inflows of asylum seekers have played in the development of restrictive policies on asylum seekers. As Meyers summarized, ‘the timing of the restrictions on asylum seekers’ have been understood to ‘better correlate with the number of applications for asylum than with the state of the economy’ (Meyers, 2004, p. 115).
With placing the increasing inflow of asylum seekers as the most important motivation for inventing restrictive policies on asylum seekers, this literature did not take into account political factors. For example, Hatton (2004), the only quantitative cross-country study on asylum policies to date, performed a quantitative analysis on the asylum policies in European countries from 1980 to 1999. Although he found the level of restrictiveness of asylum policies in a country is determined by the inflow of asylum seekers to the country as well as to other European countries, economic conditions and the degree of restrictiveness of asylum policies of other countries, no political factor was considered in his study.
There have been some exceptions to these studies. Pioneering studies that encompassed political features include Kaye (1994, 1999), which analyzed the development of British asylum policies and suggested political parties have been central to their development. Political parties, particularly the Conservative Party, placed the asylum issue on the political agenda by introducing the issue in parliamentary debates, election manifestos and party conferences. They also made use of the asylum issue as electoral strategies by introducing restrictive policies and attempting to create an image of ‘a better party’ to deal with the asylum issue. In addition, Schuster (2003) presented a broader explanation of the political backlash resulting from asylum policies. She showed how the asylum problem had yielded threats to the welfare state, liberal state and national identity in European countries. She argued that the multidimensional threats had led the countries to impose more restrictive controls on asylum seekers.
Thus, case studies, as well as other primary evidence such as survey data and parliamentary debate records, as was suggested in the beginning of this article, imply that asylum policies have not been made without any political input. In other words, the creation of restrictive asylum policies should not be understood as being solely a natural consequence of the increasing inflows of asylum seekers. If one believes that the launch of restrictive asylum polices was not simply a spontaneous response, political aspects must then be taken into account. One can, and needs to, ask the question of how political factors affect the creation of and changes in asylum policies.
Therefore, different from many studies that did not consider political aspects, this article studies how political factors determined the introduction of restrictive asylum policies. This article is also different from previous studies that incorporated political features in some respect. First, this article executes a cross-country assessment of political factors for restrictive asylum policies instead of carrying out a case study. Second, this article mainly employs quantitative data analyses using time-series cross-section (TSCS) data. Third, this article proposes, in addition to a partisanship effect, a new factor for the launch of restrictive asylum policies that was not discussed by any prior literature: election timing. Finally, this article focuses on one specific kind of policy: those that regulate asylum seekers’ economic welfare benefits. Refugee inflow raised people's concerns regarding the fiscal burden they would have to shoulder for the sake of asylum seekers. Because asylum seekers did not pay taxes while also obtaining welfare benefits while their asylum applications were in process, they were net welfare beneficiaries. As the number of asylum seekers flowing into European countries increased and the fiscal burden became heavier, the countries began formulating new asylum policies meant to reduce asylum seekers’ welfare benefits or restrict their welfare eligibility.
This article also focuses only on policies regarding asylum seekers’ welfare benefits because regulations on these benefits were developed by individual countries while most of the other major policies, such as visa imposition, ‘safe third country’ clauses and carriers’ liability, were developed through intergovernmental policy coordination, such as the 1990 Schengen Convention and the 1990 Dublin Convention. Thus, the analysis in this article assumes that domestic political factors, such as government partisanship and election timing, have more substantial effects on welfare policies than on other asylum-related policies.
This article suggests political features as important factors for the introduction of policies that restrict asylum seekers’ welfare benefits and rights. Regarding political features, this article hypothesizes that government partisanship, election timing and the interaction between the two determine the launch of restrictive policies. This article first borrows arguments from studies on migrant policies. Several studies on migrant policies showed that right-wing and left-wing parties took different positions on these policies. These studies suggest two reasons for the different attitudes. First, Lahav (2004) and Ireland (2004) showed that ideology mattered. According to their studies, left-wing party members are more committed to cultural pluralism and political, economic and social equality. Therefore, they are opposed to discrimination against migrants. In contrast, an obligation toward obedience to laws, social stability and nationalism drives right-wing parties to maintain tough attitudes toward migrants and to support restrictive migrant policies.
Second, left-wing parties consider migrants as potential supporters more than right-wing parties do (Faist, 1994; Money, 1999a). Messina (2007) found that ethnic minorities are more likely to vote for left-wing parties. Because left-wing parties perceive that migrants will eventually vote for them when they are eligible to do so, they try to provide more political and economic opportunities to them.
Whether the causal logic lies within political capital or party ideology, the literature has found evidence regarding the role of partisanship in migrant policies. This argument was tested quantitatively by Givens and Luedtke (2005). Using their own data on three European countries from 1990 to 2002, they found that although partisanship did not have an effect on policies that regulate the inflow of migrants, it brought about different outcomes regarding policies that control the political, economic and social conditions of migrants who already reside in host countries. Right-wing parties have urged for and created more restrictive migrant policies.
Left-wing parties also have legitimate reasons for being more restrictive than expected. Left-wing parties might be constrained by workers who dislike foreign people as a result of employment competition (Perlmutter, 1996). In addition, Bale et al (2010) found that left-wing parties respond to challenges from rising populist, extreme right-wing parties by taking tougher positions on migration issues. This article does not deny the rationales for left-wing parties to be restrictive on foreign people. However, based on previous literature on party politics and migration, this article anticipates that right-wing parties’ motivation outweighs left-wing parties’ motivation.
Taking a tough position on foreign people has been a popular and successful strategic choice for right-wing parties in Europe. They turned the issue of foreign people into a political one and utilized it for electoral strategies. They sometimes fueled negative public opinions regarding foreign people, showed their toughness on the issue and tried to create an image as a better party to manage the issue (Kaye, 1994, 1999; Thränhardt, 1995). Although the rationales for the partisan differences are developed in the studies of migration broadly defined, this article presumes they can be applied to the study of political asylum, too. Left-wing parties’ ideological commitment and strategic incorporation of foreign people into their constituencies have also been observed regarding asylum seekers (Schuster, 2003; Messina, 2007). Thus, this article hypothesizes that right-wing parties are more likely to introduce restrictive policies on asylum seekers’ welfare benefits than left-wing parties.Footnote 5
Right-wing governments are more likely to initiate policies that restrict and reduce asylum seekers’ welfare benefits.
Elections provide opportunities for policymakers to ‘time’ policies, and political debates concerning them, in a politically advantageous manner, so the incumbent parties can present their most popular ideas and proposals and try to create positive social and economic effects shortly before elections. Conversely, economic costs and other undesirable consequences can be timed to occur at times further from elections. Various theories and studies address this issue. For instance, the political business cycle theory argues that governments manipulate fiscal and monetary policies and try to boost the economy before elections.Footnote 6
For the same reason that popular proposals are put on the table before elections, unpopular policies are unlikely to be enacted before elections. Mikesell (1978) showed that tax increases tended to be implemented in non-election periods in the United States. Because policymakers care about the impacts of their policies, particularly on electoral outcomes, people's preferences and opinions on a specific issue may limit and control policymakers’ policy choices. Boyd (1972) found they did so particularly when the policies regulated citizens’ immediate and direct economic and social conditions, such as public order and race issues.
What were European people's attitudes and policy preferences on asylum seekers in the 1990s? The Eurobarometer survey series asked the question, ‘what should be done here in the European Community (or European Union) on the people suffering from human rights violations in their country and seeking political asylum’. More than half of the respondents said they should be accepted, but not without restriction (Table 1). The survey in 1997 also asked if the right to asylum in their countries should be easier to obtain, and 71 per cent of the respondents disagreed. Thus, the survey implies that though European people thought people seeking political asylum should be accepted by the European countries, they wanted to keep the asylum process restrictive.
Imminent elections increase the pressure on political parties to not deviate from core public opinions. As the inflow of asylum seekers rises, the tolerance of native people lessens. Negative attitudes toward asylum seekers sometimes result in violent attacks on them. Consequently, the number of destructive actions against asylum seekers has increased in many European countries, particularly during the ‘asylum crisis’ in the early 1990s.Footnote 7
In many cases, it seems that politicians knew their parties could be rewarded or punished by voters in elections based on what the parties had done on issues related to asylum seekers in the past (Saggar, 1997; Bosswick, 2000). As the political business cycle theory and other studies on the timing of preferred and disliked policies imply, upcoming elections are believed to have provided opportunities for or pressure on political parties to introduce restrictive asylum policies because public opinion was negative in prior decades. Thus, this article hypothesizes that the likelihood of the introduction of restrictive asylum policies increases during pre-election periods.
Restrictive policies on asylum seekers’ welfare benefits are more likely to be introduced before elections.
If elections provide incentives and motivation for political parties to be more restrictive on asylum seekers, do they have the same effect on both left-wing and right-wing parties? This article hypothesizes they do not, because they provide different incentives for the two types of parties to introduce restrictive policies. The literature suggests that right-wing parties have actually tried to politicize and mobilize issues on asylum seekers, as well as foreign people in general, and connect them with other broader issues like crime, unemployment and welfare dependency in the hopes that the strategy would help the parties gain more electoral support (Bale, 2003). For example, in France, right-wing parties were criticized by left-wing parties that they used asylum seekers as the ‘scapegoat’ during an economic crisis (Soltesz, 1995, p. 273).
For right-wing parties, asylum issues and negative public opinion on them can be a ‘reliable weapon in competitive electoral systems’ (Thränhardt, 1995). In addition, Bale (2003) concluded that many center-right parties try to expand a right bloc in party competition with left-wing parties by acknowledging the stances of extreme right-wing parties regarding foreign people and helping them to siphon off votes from the left-wing bloc. Because restrictive positions on asylum issues are compatible with their original ideological commitment to laws, social stability and nationalism, there is little reason for right-wing parties to hesitate to utilize the asylum issue in election campaigns when public attitudes toward asylum seekers are negative.
By contrast, even in the face of pressure from public opinion and upcoming elections, a shift in policy positions on foreign people is believed to be a last resort for left-wing parties. As Bale et al (2010) showed, a policy-shift strategy (adopt) is used only when other strategies like sustaining their own stances (hold) or refuting the salience or relevance of issues on foreign people as electoral issues (defuse) do not work out any more. They do not shift their positions until they anticipate a loss of votes to extreme right-wing parties and expect the formation of centre-right governments. For example, the withdrawal of automatic entitlement of asylum seekers to federal support in 1991 by the Austrian Socialist Party was a reluctant decision due to increasing public hatred of asylum seekers. The decision was made to overcome a popularity deficit, prevent further growth of Haider's Austrian Freedom Party and avoid electoral defeat (Amrute and Pfohman, 2001).
Elections provide incentives for political parties to become more restrictive on asylum seekers by providing opportunities to make use of their tough positions or by making it harder to sustain their soft positions. Because of the different incentives for left-wing and right-wing parties, this article hypothesizes that right-wing parties become more likely to introduce restrictive policies than left-wing parties because left-wing parties are more reluctant to do so due to their ideological commitment and electoral constituencies.
The electoral effect is greater for right-wing parties than for left-wing ones.
Data and Variables
This article employs part of the data constructed by Hatton (2004) on European asylum policies. He categorized 14 different asylum policies and coded when the policies were created and/or abolished. Among the policies categorized in the study, those concerning welfare benefit restrictions are used as an indicator of curtailing asylum seekers’ economic welfare. The data cover 13 European countries from 1981 to 2000. The main dependent variable is the introduction of policies that restrict or reduce asylum seekers’ welfare benefits.Footnote 8 Thus, either introducing a new restrictive policy or modifying a policy to be more restrictive is coded as 1, while no such policy change is coded 0.
Of course, such binary coding has limitations; for example, it cannot indicate the magnitude of a policy change. As an extreme example, a total withdrawal of welfare benefits and a slight reduction of benefit levels are treated equally in such coding, when obviously they are not. However, measuring asylum or migration policies with continuous values is almost impossible (Money, 1999b). In addition, this article is interested not in the magnitude of the policy changes but in whether an asylum policy was modified. Therefore, binary coding is used despite its limitations.
The factors that receive the most attention are government partisanship and election timing. Government partisanship can be measured in several ways. One way of measuring partisanship is the degree to which parties lean toward the left or right. Another is to calculate the percentage of left-wing or right-wing parties in Parliament or in a cabinet (Iversen and Cusack, 2000). This article adopts both approaches and uses two variables for government partisanship. For the first approach, the normalized average of four distinct partisanship scores is used (Castles and Mair, 1984; Laver and Hunt, 1992; Warwick, 1994; Huber and Inglehart, 1995, left1). For the second, the percentage of left portfolios in a cabinet is used, from Swank's Comparative Parties database (left2). The correlation between the two is 0.8392.
Two variables are used for election timing. This article focuses on parliamentary elections because most of the countries in the analysis have a parliamentary system. A variable is coded as 1 for a year previous to a parliamentary election (election1). This may be a very crude measurement. The mechanical coding cannot guarantee that the policy change was made for the purpose of an electoral campaign. Because election timing is endogenous in many European countries, an asylum policy change might be made without the expectation of an upcoming election.Footnote 9 Also, debates for some policy changes last for very long periods, sometimes more than a year. In those cases, it is quite difficult to capture the real timing of policy changes. To supplement the variable, a calculation was made of how many years there were between an asylum policy and the next election: this was used as another measurement of election timing (election2).Footnote 10 It is hypothesized that as the number of years to the next election decreases, the likelihood of policy introduction increases.Footnote 11
The analyses begin with replicating the models in Hatton (2004) regarding the asylum policy level. Thus, the policy level and policy introduction of other countries, the asylum seekers’ inflow to a country as well as to other European countries, and the change in GDP per capita are controlled. The inflow data are from the OECD's Trends in International Migration. The variable is the percentage of the total population. GDP per capita data are from the World Bank's World Development Indicators.
This article adds more control variables that are expected to have an effect on either policy levels or policy introduction. Bale et al (2010) suggest that the rise of extreme right-wing parties and their politicization of issues on foreign people constrained other parties not to pursue liberal policies. To control for the rise of extreme right-wing parties, the models include their vote share in elections. This article follows Kitschelt (1995) and Mudde (2007) for the list of extreme right-wing parties and expects that the large vote share by extreme right-wing parties raises the likelihood of policy introduction.
This article also expects that the extreme right-wing party variable can capture the public opinion effect at least partially within each country. Although negative public opinions on asylum seekers is believed to give pressure for restrictive asylum policies, the lack of consistent and reliable time-series cross-country surveys on asylum issues makes the inclusion of public opinion variables difficult. Because support for parties is correlated with negative public attitudes toward foreign people including asylum seekers (Lubbers et al, 2002), the variable will indirectly indicate the change of public opinion within each country over time.
The analysis also includes a variable of the number of veto players in the political decision-making process. Veto player theory suggests it is hard to change the status quo substantially if there are many veto players because the presence of many veto players reduces the size of the winset, that is, a set of alternatives that can defeat the status quo (Tsebelis, 2002). This implies that the number of veto players decreases the likelihood of policy introduction because the introduction of a restrictive asylum policy indicates a change in the status quo regarding asylum issues. Data on the number of veto players are from Tsebelis’ website.
The restrictions on welfare benefits began to be discussed because of people's concern regarding the fiscal burden placed on them by asylum seekers (Geddes, 2003). Thus, countries suffering from budget deficits will have stronger incentives to cut welfare support for asylum seekers. Therefore, this analysis also controls for the balance of government budgets. The data are from IMF, International Financial Statistics.
As was discussed before, many of the refugee policies of European countries have been coordinated through inter-governmental agreements, such as the 1990 Schengen Convention and the 1990 Dublin Convention, though welfare policies have been adopted more independently than other policies (Gibney and Hansen, 2003). To control for such spillover effects, whether a policy on welfare benefits was adopted by other European countries in the same or previous years is also controlled.
The countries that already have restrictions on welfare benefits to asylum seekers are expected to have weak incentives to introduce additional restrictions. In other words, the current level of policies on asylum seekers’ welfare benefits matters for policy introduction. For this reason, the policy level from Hatton (2004) is also included in the model. Finally, the current number (stock) of refugees, in addition to the inflow of asylum seekers, is controlled. The data are from the United Nations High Commissioner for Refugees, Statistical Yearbook. The variable is calculated as the percentage of the total population. Table 2 provides a description of the data.
Before examining the introduction of policies that regulate asylum seekers’ welfare benefits and rights, this article replicated the models on asylum policy levels in Hatton (2004) using a time-series cross-sectional OLS model. Model (1) in Table 3 examined all the asylum policies and model (3) used only policies on welfare benefits. The results are quite similar with those in Hatton (2004, p. 31). Models (2) and (4) added variables that are introduced in this article. The variable for other countries’ policy levels has a positive coefficient: the result shows that the policy levels of European countries moved together even after asylum seekers’ inflow was controlled. This implies policy coordination between the countries (Gibney and Hansen, 2003). Also, the inflow of asylum seekers, particularly the overall inflow to European countries, raised the restrictiveness of asylum policies in each country. Economic downturns, as Hatton (2004) also found, make the policies more restrictive.
Government partisanship and income levels also have an effect on policy levels, particularly on overall policy levels. Right-wing governments tend to maintain more restrictive policies. The richer the countries, the more liberal their asylum policies. Government partisanship may have an effect on the level of overall asylum policies for the same reason the hypothesis of this article suggests. The wealth of a country also matters because it provides financial resources for taking care of asylum seekers. However, what this article is interested in is not the policy level but the incident of policy introduction. Thus, this article moves on to the analysis of policy introduction.
Because the policy introduction data are time-series cross-sectional data with a binary dependent variable, a cross-sectional time-series probit model is used, one that is based on Beck et al (1998). The government partisanship variables and the election variables test the first and second hypotheses. To test the third hypothesis, the model includes an interaction term between the two variables. Table 4 presents the results of the analysis.
The results demonstrate, first of all, that the pre-election variable is statistically significant, with expected coefficients. Policies restricting and reducing asylum seekers’ welfare are more likely to be introduced in the year immediately before a general election (election1). The likelihood of policy introduction also increases as the next election is approaching (election2). The partisanship variables are statistically significant, too. Right-wing governments are more likely to introduce restrictive asylum policies particularly when elections are in the near future.Footnote 12 As was seen in the analysis on policy level, income levels of countries also decide the likelihood of policy introduction. Finally, the current level of policies on asylum seekers’ welfare benefits matters: when a country lacks such a policy, the country is likely to initiate a new one.
Table 5 presents the results of robustness checks. As was discussed earlier, many of the European countries have endogenous election timing. Thus, model (1) counts only elections that were not called by governments, but held by election rules.Footnote 13 Thus, the variable election3 indicates a year previous to a general election that was mandated by rules. Then, election timing in model (1) can be considered exogenous. Model (2) includes only countries where restrictive policies were introduced at least once during the analysis period. Some European countries could avoid the asylum crisis: not many asylum seekers flowed to Southern European countries because they did not have economic merits. Scandinavian countries, except for Sweden, suffered from the inflow of asylum seekers less than continental European countries did because of their geographical location. At the same time, left-wing governments prevailed in many of these countries in the 1980s–1990s. Although the inflow and stock levels of asylum seekers are controlled for in each analysis, the inclusion of these countries may produce biased results. Thus, countries that did not introduce restrictive policies on asylum seekers’ welfare assistance are excluded in model (2). However, the results are not substantially changed.
The GDP growth rate was included in Hatton's model, but excluded in Table 4 due to multicollinearity because the budget deficit variable has the GDP growth rate as a denominator. Model (3) includes it, but the inclusion does not change the result much. Model (4) excludes the asylum seeker inflow variable because it results in a multicollinearity problem with the other two asylum seeker variables: the stock of refugees in each country and the inflow of asylum seekers to other European countries. The exclusion makes the refugee stock variable statistically significant: a large stock of refugees increases the likelihood of policy introduction. Thus, we cannot deny that the number of asylum seekers, particularly the number of those who already resided in the country, has put pressure on the host country. Finally, model (5) uses the budget deficits of local governments because the assistance to asylum seekers has usually been funded by local governments (Bloch, 2000). The budget deficit variable, however, is still statistically insignificant.
The interaction terms in Tables 4 and 5 have expected coefficients, but some of them are not statistically significant. However, interaction terms are often insignificant due to multicollinearity. With that in mind, a graphic illustration of the interacting relationship between government partisanship and election periods is presented in Figures 1 and 2.Footnote 14
The graphs in Figure 1 show how the effect of government partisanship depends on election timing, and show that right-wing and left-wing parties differ only in pre-election periods. The top two graphs demonstrate that the partisanship variables are statistically significant with negative coefficients only in pre-election periods. The bottom two graphs tell us that the partisanship variable is not statistically significant when elections are remote, but it becomes significant with a negative coefficient as elections are approaching. The partisan difference appears only in the years immediately before elections. Thus, the graphs imply that significant partisan differences do not appear when the next election is remote. However, when an election is upcoming, right-wing parties are more likely to introduce restrictive policies than are left-wing parties.
The graphs in Figure 2 demonstrate how the pre-election effect differs between left-wing and right-wing parties. The top two graphs show that the variable election1 has an effect only when the partisanship variables have small values: in other words, when there are right-wing governments. Under left-wing governments, the variable is not statistically significant. Similarly, the bottom graphs tell us that the variable election2 has an effect only under right-wing governments. The graphs imply that the pre-election effect operates only among right-wing parties. Election periods do not matter for policy introduction likelihood for left-wing parties. In other words, left-wing parties are not affected by upcoming elections.
In sum, elections drive political parties to promote and act on restrictive asylum policies, on the basis of negative public attitudes toward asylum seekers. Elections may drive left-wing parties as well as right-wing ones to do so. However, the overall effect of election timing seems to be driven by right-wing parties. The results of the analysis above imply that while the probability that left-wing governments will introduce restrictive policies increases only by 0.03 from a non-election period to an election period, the probability of right-wing governments doing so increases by 0.52. While the policy shift is believed to be a last resort for left-wing parties because of their ideological commitment and electoral constituency (Schuster and Solomos, 2004; Bale et al, 2010), it has been considered as a ‘reliable weapon in competitive electoral systems’ for right-wing parties (Thränhardt, 1995). Therefore, though there is no significant partisan difference in non-election periods, right-wing parties pass restrictive asylum policies more than left-wing parties do in election periods.
This article began by questioning whether restrictive asylum policies were merely spontaneous responses to the rising level of asylum seekers’ inflow or outcomes determined by political factors. The results from the quantitative data analysis in this article imply that restrictive asylum polices are not just spontaneous responses to the influx of asylum seekers but are the result of political determinants. The introduction of policies that restrict asylum seekers’ welfare benefits was driven by political factors such as government partisanship and election timing as well as by the (stock) level of asylum seekers and economic factors such as the level of national income. The results suggest that while the increasing level of asylum seekers might have provided an incentive to form new devices to restrict further inflows, the asylum issue was politicized, mobilized and utilized by political, particularly right-wing, parties.
The results also imply that partisan differences on foreign people may not be stationary. People often take a static approach in explaining differences between left-wing and right-wing parties with regard to policies regarding foreign people, such as whether left-wing parties are pro-migration or not. The literature, including this article, implies, however, that party politics may have more complicated dynamics. Political parties’ policies and positions on the issues are influenced by their internal constituency structure (Perlmutter, 1996), the strategies of their opponent parties (Bale et al, 2010), or their positions and roles in the party system (Adams et al, 2006). Many factors, in addition to their overall ideological stances, can provide opportunities or pressure for parties to shift their positions in either a more liberal or more restrictive direction. Therefore, a simple left/right division may not be able to explain sophisticated differences between political parties, particularly regarding new political issues like migration as well as European integration and ecology (Hooghe et al, 2002). Therefore, more comprehensive approaches will be needed to understand the multi-dimensional aspects of party politics and migration policies.
Asylum policies in Europe were ‘Europeanized’ after the Treaty of Amsterdam in 1999. Consequently, the asylum policies of European countries are no more an exclusive prerogative of individual states. As a result, the study of domestic political factors such as government partisanship and election timing might not be relevant any more. However, the approach of party politics is a valuable research tool for European Union politics as well as for domestic politics (Hix, 2008). Therefore, understanding the dynamics of party politics at the country level will also help to comprehend European Union politics and policymaking.
Hansard, 21 January 1992, Column 273–274.
In 1985, about 170 000 people went to European countries in search of political asylum. The number jumped to 430 000 in 1990 and peaked at 700 000 in 1992 (OECD, Trends in International Migration, various years).
Several Eurobarometer surveys asked respondents whether asylum seekers should be accepted with or without restriction, or not to accept them. Whether the average answer of people who voted for center-left parties is different from that of people voting for center-right parties was tested using the 1993 survey, and it turned out the responses between the two groups of people are statistically different at the 0.05 significance level. More details on the survey questions will be presented later in this article.
The countries are Austria, Belgium, Denmark, Finland, France, Germany, Ireland, Italy, the Netherlands, Portugal, Spain, Sweden and the United Kingdom.
This article does not attempt to make a dichotomous categorization of left-wing and right-wing governments. Methodologically, this article employs average partisanship scores of government parties and the legislative seat share of the parties, and uses them as indicators of government partisanship. Conceptually, however, this article suggests that governments in which mainstream left-wing (right-wing) parties play a leading role are considered left-wing (right-wing) governments though we can find atypical cases, such as the grand coalition between the Social Democratic Party and the Christian Democratic Union in Germany after 2005.
Since the early works on the political business cycle theory in the 1970s, the theory has been challenged and revised by many scholars. They modified or questioned it with the prospective and/or rational voter theory, the partisan model and the surfing theory, among other constructions. However, the basic idea of incumbents’ incentive to push for popular policies before elections has survived through the series of modifications. See Alesina et al (1997) and Drazen (2000) for surveys and tests on diverse political business cycle models.
For example, the number of violent terrorist attacks against asylum seekers more than doubled in the early 1990s in Scandinavian countries (Björgo, 1993).
I also consulted Zetter et al (2003) to ensure the validity of Hatton's data and added three more cases. The three cases are as follows. The Netherlands in 1987 passed the Regulation on the Reception of Asylum Seekers and excluded asylum seekers from the National Assistance system. In 1997, Germany modified its 1993 policy that had withdrawn cash subsidies for the first year of application. The 1997 law removed the ‘first year’ clause and extended the in-kind benefits to the entire period of application. Finally, Britain in 1999 replaced cash benefits with the voucher system.
However, more than 70 per cent of elections in my data were held regularly, every 4 or 5 years, depending on electoral rules, particularly in the 1990s. This implies that most of the elections were actually exogenous, despite official rules.
To deal with the endogeneity of election timing, Williams (2010) calculated the probability of holding elections and used it, not real election timing, as a predictor of engaging in military conflicts. However, it also shows that the probability increases as time goes on since the previous election and as the next election is approaching.
Election years are hard to code. Strictly speaking, each day before an election day should be coded as 0, while all the days after the election day should be coded with large numbers depending on the number of years between elections. Thus, it is technically impossible to code election years. A choice was made to suffer from the missing data problem rather than from bias and leave the election years as missing data.
Scandinavian countries have exogenous electoral timing. Other European countries also have maximum terms for their parliaments. If the government does not call an election within the term, it should hold an election in the last year of the term.
The circles and the solid lines in the graphs indicate the coefficients of the partisanship variables in Figure 1 and those of the pre-election variables in Figure 2. The x-marks and the dotted lines indicate the 95 per cent confidence intervals of the variables. For example, the top-left graph in Figure 1 shows that the variable left1 is not statistically significant in non-election periods, but it is significant with a negative coefficient in election periods.
Adams, J., Clark, M., Ezrow, L. and Glasgow, G. (2006) Are niche parties fundamentally different from mainstream parties? The causes and the electoral consequences of Western European parties’ policy shifts, 1976–1998. American Journal of Political Science 50 (3): 513–529.
Alesina, A., Roubini, N. and Cohen, G.D. (1997) Political Cycles and the Macroeconomy. Cambridge, MA: MIT Press.
Amrute, S. and Pfohman, S. (2001) Country study: Austria. In: European Refugee Fund Community Actions (ed.) Decentralisation of Asylum: Refugee Reception Procedures in the European Union. Berlin: Berlin Institute of Comparative Social Research.
Bale, T. (2003) Cinderella and her ugly sisters: The mainstream and extreme right in Europe's bipolarising party systems. West European Politics 26 (3): 67–90.
Bale, T., Green-Pedersen, C., Krouwel, A., Luther, K.R. and Sitter, N. (2010) If you can’t beat them, join them? Explaining social democratic responses to the challenge from the populist radical right in Western Europe. Political Studies 58 (3): 410–426.
Beck, N., Katz, J.N. and Tucker, R. (1998) Taking time seriously: Time-series-cross-section analysis with a binary dependent variable. American Journal of Political Science 42 (4): 1260–1288.
Björgo, T. (1993) Terrorist violence against immigrants and refugees in Scandinavia: Patterns and motives. In: T. Björge and R. Witte (eds.) Racist Violence in Europe. New York: St. Martin's Press.
Bloch, A. (2000) A new era or more of the same? Asylum policy in the UK. Journal of Refugee Studies 13 (1): 29–42.
Boswell, C. (2003) European Migration Policies in Flux: Changing Patterns of Inclusion and Exclusion. Malden, MA: Blackwell Publishing.
Bosswick, W. (2000) Development of asylum policy in Germany. Journal of Refugee Studies 13 (1): 43–60.
Boyd, R.W. (1972) Popular control of public policy: A normal vote analysis of the 1968 election. American Political Science Review 66 (2): 429–449.
Castles, F.G. and Mair, P. (1984) Left right political scales: Some expert judgments. European Journal of Political Research 12 (1): 73–88.
Crisp, J. (2003) Refugees and the global politics of asylum. In: S. Spencer (ed.) The Politics of Migration: Managing Opportunity, Conflict and Change. Malden, MA: Blackwell Publishing.
Drazen, A. (2000) The political business cycle after 25 years. NBER Macroeconomics Annual 15: 75–117.
Faist, T. (1994) How to define a foreigner? The symbolic politics of immigration in German partisan discourse, 1978–1992. West European Politics 17 (2): 50–71.
Geddes, A. (2003) The Politics of Migration and Immigration in Europe. London: Sage Publications.
Gibney, M.J. and Hansen, R. (2003) Asylum Policy in the West: Past Trends, Future Possibilities. WIDER Discussion Paper No. 2003/68.
Givens, T. and Luedtke, A. (2005) European immigration policies in comparative perspective: Issue salience, partisanship and immigrant rights. Comparative European Politics 3 (1): 1–22.
Hatton, T.J. (2004) Seeking asylum in Europe. Economic Policy 38 (1): 5–62.
Hix, S. (2008) Towards a partisan theory of EU politics. Journal of European Public Policy 15 (8): 1254–1265.
Hooghe, L., Marks, G. and Wilson, C.J. (2002) Does left/right structure party positions on European integration? Comparative Political Studies 35 (8): 965–989.
Huber, J. and Inglehart, R. (1995) Expert interpretations of party space and party locations in 42 societies. Party Politics 1 (1): 73–111.
Ireland, P.R. (2004) Becoming Europe: Immigration, Integration, and the Welfare State. Pittsburgh, PA: University of Pittsburgh Press.
Iversen, T. and Cusack, T.R. (2000) The causes of welfare state expansion: Deindustrialization or globalization? World Politics 52 (3): 313–349.
Kaye, R. (1994) Defining the agenda: British refugee policy and the role of parties. Journal of Refugee Studies 7 (2/3): 144–159.
Kaye, R. (1999) The politics of exclusion: The withdrawal of social welfare benefits from asylum seekers in the UK. Contemporary Politics 5 (1): 25–45.
Kitschelt, H. (1995) The Radical Right in Western Europe: A Comparative Analysis. Ann Arbor, MI: The University of Michigan Press.
Lahav, G. (2004) Immigration and Politics in the New Europe: Reinventing Borders. Cambridge, MA: Cambridge University Press.
Laver, M. and Hunt, W.B. (1992) Policy and Party Competition. New York: Routledge.
Lubbers, M., Gijsberts, M. and Scheepers, P. (2002) Extreme right-wing voting in Western Europe. European Journal of Political Research 41 (3): 345–378.
Messina, A.M. (2007) The Logics and Politics of Post-WWII Migration to Western Europe. Cambridge, MA: Cambridge University Press.
Meyers, E. (2004) International Immigration Policy: A Theoretical and Comparative Analysis. New York: Palgrave Macmillan.
Mikesell, J.L. (1978) Election periods and state tax policy cycles. Public Choice 33 (3): 99–106.
Money, J. (1999a) Defining immigration policy: Inventory, quantitative referents, and empirical regularities. Presented at the 95th Annual Meeting of the American Political Science Association, Atlanta, Georgia.
Money, J. (1999b) Fences and Neighbors: The Political Geography of Immigration Control. Ithaca, NY: Cornell University Press.
Mudde, C. (2007) Populist Radical Right Parties in Europe. Cambridge, MA: Cambridge University Press.
Perlmutter, T. (1996) Bringing parties back in: Comments on ‘modes of immigration politics in liberal democratic societies’. International Migration Review 30 (1): 375–388.
Saggar, S. (1997) The dog that didn’t bark? Immigration, race and the election. In: A. Geddes and J. Tonge (eds.) Labour's Landslide: The British General Election 1997. Manchester, UK: Manchester University Press.
Schuster, L. (2003) The Use and Abuse of Political Asylum in Britain and Germany. London: Frank Cass.
Schuster, L. and Solomos, J. (2004) Race, immigration and asylum: New Labour's agenda and its consequences. Ethnicities 4 (2): 267–300.
Soltesz, S. (1995) Implications of the Conseil Constitutionnel's immigration and asylum decision of August 1993. Boston College International and Comparative Law Review 18 (1): 265–315.
Steiner, N. (2000) Arguing About Asylum: The Complexity of Refugee Debates in Europe. New York: St. Martin's Press.
Thränhardt, D. (1995) The political uses of xenophobia in England, France and Germany. Party Politics 1 (3): 323–345.
Tsebelis, G. (2002) Veto Players: How Political Institutions Work. New York/Princeton, NJ: Princeton University Press.
Warwick, P. (1994) Government Survival in Parliamentary Democracies. Cambridge, MA: Cambridge University Press.
Williams, L.K. (2010) Endogenous election timing, election cycles and international conflict. Presented at the 106th Annual Meeting of the American Political Science Association, Washington DC.
Zetter, R., Griffiths, D., Ferretti, S. and Pearl, M. (2003) An Assessment of the Impact of Asylum Policies in Europe 1990–2000. Part 2. Home Office Online Report 17/03. London: Home Office.
About this article
Cite this article
Han, K. Political use of asylum policies: The effects of partisanship and election timing on government policies regarding asylum seekers’ welfare benefits. Comp Eur Polit 11, 383–405 (2013). https://doi.org/10.1057/cep.2012.24
- party politics