Party Representatives’ Adaptation to Election Results: Dyadic Responsiveness Revisited

Abstract

Politicians’ dual responsibilities to respect their party and also be responsive to their constituents is surprisingly lacking in studies of representation. How do politicians—especially those who function in strong-party systems—individually respond to their constituents’ preferences? We make use of an original, large-scale survey of politicians and the recent success of the Sweden Democrats in the elections in Sweden to show that important adaptation takes place within the party structure. Individual politicians are responsive to signals about voters’ preferences, and they act on these signals by internally lobbying their party leaders to change the party’s positions in the direction of their constituents’ preferences. These results provide a rationale for why niche parties invest in elections even if they are unlikely to enter government: Their electoral successes can cause change in other parties. The results also add a new angle to the discussion of how anti-immigration parties affect mainstream parties, a hotly debated issue in many advanced democracies.

Keywords

European politics responsiveness immigration Sweden political parties

How do election results affect individual-level, political responsiveness in systems with strong political parties? Studying the responsiveness of individual politicians in strong-party systems is important because it allows us to learn how politicians handle their role as dual agents for both their party and their constituents. Previous work clearly shows that while individual politicians may desire to be responsive to voters, they also have incentives to support their party’s leadership (Carey & Shugart, 1995; Preece, 2014). Indeed, in systems with strong political parties, individual politicians generally take policy positions in conjunction with their parties’ official policies (e.g., Carey, 2007, 2009; Strøm, 1997).

We study how individual politicians respond to the electoral success of other parties because this affords us an opportunity to see how they react when their incentives to be responsive to voters conflicts with the positions of their party. Significantly, election results can provide politicians information about voters’ issue preferences, especially in multiparty systems where parties are often associated with specific policies (Adams, 2012; Budge, 1994; Somer-Topcu, 2009). Therefore, when voters suddenly disagree with a party and start voting for other parties, representatives of that constituency need to handle conflicting incentives: They will want to respond to these voters’ policy wishes and at the same time stay loyal to the party position.

We argue that politicians handle these dual pressures by lobbying party leaders to change policy in the direction of their constituents’ positions. Indeed, because the party’s position determines policy outcomes (Gauja, 2013), internally advocating for change in the party’s position is the mechanism by which politicians can act to be individually responsive to voters. Parties’ policies should therefore be expected to change as individual politicians respond to expressions of their constituents’ preferences. In this article, we test how other parties’ electoral success affects the degree to which politicians lobby their party leaders to change the party’s position.

While the study of individual politicians’ response to public opinion has long tradition in political science (Golder & Stramski, 2010; Mayhew, 1974; Miller & Stokes, 1963; Powell, 2000), not much attention has been given to the work that politicians put into changing their party’s position from within. Instead, researchers have focused on legislators’ public actions (Stratmann, 2000). To better understand how politicians in strong-party systems are individually responsive to election results in their constituencies, we study what happens informally inside the political party. We provide insights into this issue by collecting unique survey data from Swedish municipal politicians in the wake of the 2014 Swedish election.

For reasons we outline later, the Swedish case provides an ideal opportunity to study individual politicians’ responsiveness to election results. Significantly, the 2014 Swedish elections provide a case where a niche, issue-oriented party (the Sweden Democrats) made large gains in some but not other areas of Sweden. Some, but not all, politicians thus received a strong signal that their constituents preferred more restrictions on immigration. We study this situation by collecting a large-scale survey of local politicians in 2015, 5 months after the election.

Our analysis provides evidence that the officials individually responded to their local election results by changing how they internally lobbied party leaders. In areas where the Sweden Democrats increased their vote share by a greater amount, municipal officials were more likely to report that they had lobbied the leaders of their party in the months between the election and the survey. Our article contributes to opening up the black box of the interaction of interparty competition and intraparty dynamics and it shows that politicians engage in intraparty, issue lobbying in efforts to be responsive to their constituents.

Effects of Election Results on Individual Politicians in Party-Centered Systems

The relationship between constituents’ preferences and politicians’ actions are at the heart of democratic theory (e.g., Lijphart, 1984; Monroe, 1979; Powell, 2000; Stimson, MacKuen, & Erikson, 1995, for an overview, see Wlezien & Soroka, 2007). Classic theories about democratic accountability suggest that this relationship holds because politicians are held accountable for their actions (Downs, 1957). Put simply, constituents have preferences on the important issues of the day and they vote on the basis of those issues. As a result, ambitious politicians have the incentive to learn about and be responsive to public opinion. If they do not respond to their constituent’ preferences, they risk being electorally punished and losing office.

Elections are an important way in which politicians can learn about constituents’ preferences. Classic models of voting behavior assumes that citizens vote for the party/candidate closest to their own position, which means that elections provide information about voters’ preferences. An increase in a given party’s vote share indicates that citizens have preferences closer to that party than to the competition (Budge, 1994; Somer-Topcu, 2009; Somer-Topcu & Zar, 2014). While there are other signals about voters’ preferences (e.g., opinion polls, protests, etc.), elections are particularly important because politicians who care about winning reelection want to be responsive to the preferences of individuals who vote.

In multiparty systems, election results can be particularly informative by providing signals about voters’ attitudes on specific issues of the day. While much research has focused on shifts in parties’ overall left–right positions following elections (Adams, Clark, Ezrow, & Glasgow, 2006; Adams, Haupt, & Stoll, 2009; Budge, 1994; Ezrow & Hellwig, 2014), there are reasons to think that multiparty election results can be even more informative about voters’ preferences on specific issues. As Broockman (2016) argues, ideological scales do not always do a good job describing citizens’ preferences and so their votes may not always represent positions on a left–right scale (see also Adams, 2012). However, voters are likely to have positions on individual issues that in some situations are likely to drive their votes (Belanger & Meguid, 2008). This is especially true in multiparty systems with a large number of parties. When there are a larger number of parties, parties often use specific issue positions to differentiate themselves from the competition (Green-Pedersen, 2007). Niche parties, in particular, are often associated with specific issues and when those parties win more votes it provides a signal about voters’ preferences on those issues (Abou-Chadi, 2014; Bale, Green-Pedersen, Krouwel, Luther, & Sitter, 2010; van de Wardt, 2015). Indeed, evidence shows that parties as an organization react to niche party success as though it is a strong signal of voters’ preference (e.g., Adams & Somer-Topcu, 2009; Spoon, Hobolt, & de Vries, 2014; cf. Meguid, 2005, 2008).

Individual politicians in strong-party systems, however, may be unresponsive to election results because they have strong incentives to worry about the preferences of their party leaders (Öhberg & Naurin, 2015, cf. Lindstädt & Vander Wielen, 2014). In strong-party systems, party leaders are in charge of putting names forward for the party, meaning rank-and-file politicians only have a shot at being on the ballot if they please party leaders, even at the cost of being unresponsive to voters (Carey, 1996; Hazan & Rahat, 2010; Rahat & Hazan, 2001). And even when politicians have incentives to build a large personal vote, party leaders can still strongly influence the legislators’ behavior (Preece, 2014). Party-centered systems are thus often described as giving little room for the individual to act on behalf of voters.

Traditional explanations have often stopped there, focusing on politicians’ will to toe the party line, even at the expense of voter preferences. Accountability and responsiveness operate less well at the individual level, according to model such as the responsible party government model (American Political Science Association Committee on Political Parties, 1950; Downs, 1957; Klingemann, Hofferbert, & Budge, 1994; McDonald & Budge, 2005). There is no room in that story for the individual politicians to act as a personal delegate of the voters.

However, a strong focus on politicians’ incentives to toe the party line misses the potential of individual responsiveness in at least two ways. First, politicians can, should, and want to work inside the party to shape policy that is more in line with their constituents (cf. Dudzinska, Poyet, Costa, & Wessels, 2014). Research has not looked into this aspect of responsiveness. Instead, responsiveness to public opinion is measured using roll call votes (Bafumi & Herron, 2010; Butler & Nickerson, 2011; Kousser, Lewis, & Masket, 2007; Miller & Stokes, 1963), parliamentary debates (Proksch & Slapin, 2012), motion-writing (Bowler, 2010), and discussions with the media (Arceneaux, Johnson, Lindstädt, & Wielen, 2015). Such studies provide important understanding of how politicians act openly in response to public opinion. However, they miss the action that takes place within parties, such as representatives’ efforts to influence their party colleagues to behave responsively. When representatives take the parliamentary floor, write an article, or cast their vote in parliament, these are often the last acts in series of events. If we do not study legislators’ behavior within the party, we may severely underestimate the impact of elections on representation (Fagerholm, 2015; Öhberg & Naurin, 2015; cf. Budge, Ezrow, & McDonald, 2010).

Second, savvy parties should be interested in feedback from rank-and-file members about the party positions. Rank-and-file legislators are likely to invest in knowing what their constituents care about and can thus provide the leaders with insights into the public’s mood. Further, elections are a type of external stimuli that can “catch the attention of someone in the party who would see fit to argue that adaptive change would be needed in order for the party to ‘do better’ in some way than it would otherwise do” (Harmel & Janda, 1994, p. 267). So even though internal-party lobbying is not likely to be visible to voters,¹ this type of lobbying is important because it is a fundamental mechanism by which elections can improve representation. If enough party members pressure leaders for change, leaders are likely to change the party’s position. By focusing on individual politicians’ responsiveness, our study also provides micro-foundations for the literature on party response to voter signals.

Case Selection: The 2014 Swedish Election

We use the increased success of the Sweden Democrats in the 2014 Swedish municipal elections to study whether changes in election results affect how politicians work within the party structure. There are several reasons why the 2014 Swedish municipal elections is a good case for studying this question. First, politicians in Sweden have strong incentives to cater to party leaders’ preferences (Dancygier, Lindgren, Oskarsson, & Vernby, 2015). Because we are interested in studying whether politicians are responsive to their constituents despite incentives to be responsive to party leaders, it is important to choose a country where politicians have incentives to be responsive to party leaders. Sweden fits the bill. Party leaders in Sweden play an important role in deciding which candidates make it onto the ballot. These incentives contribute to high levels of party cohesiveness in Sweden (Jensen, 2000), with members of local and national assemblies tending to be loyal party representatives (Erlingsson, Kölln, & Öhberg, 2016).

Second, in the 2014 election, the Sweden Democrats made large electoral gains in many (but not all) areas. To study the role of elections in providing a signal about voters’ preferences, we need variation in that signal. The change also has to be large enough that politicians have incentives to pay attention. In the Swedish 2014 municipal election, the Sweden Democrats earned an average of 9.3% of the vote, nearly doubling the amount they had earned in the previous election, 4 years earlier (they earned an average of 4.9% in the 2010 municipal elections). Further, there was significant variation in where the Sweden Democrats picked up votes. In some cases, the Sweden Democrats did not gain any votes, while in others they increased their vote share by 18 percentage points (see Figure 1). We use this variation to study whether it is correlated with the reported lobbying behavior of individual politicians.

Figure 1.

Distribution of the Sweden Democrats’ vote share for the municipalities in our sample.

Third, and related to the second point, we study municipal officials to use the variation in the signals that politicians receive. National MPs have incentives to be responsive to the whole electorate, which means that there is no clear variation in what the relevant electorate wants. Municipal officials, on the other hand, face different electorates. At the same time, these officials are connected to the national party’s policy positions.² In fact, Swedish municipality elections always take place on the same day as the national elections, once every fourth year. National campaigns and national policy positions therefore tend to have a large impact on who wins at the local level (Berg & Oscarsson, 2012), and local campaigns are affected and often driven by national politics (Nord, 2007).³ Local politicians thus have incentives to lobby party leaders about party policy because it is part (though not all) of what affects their own political fate. In sum, local politicians have the same incentive as national politicians to lobby party leaders while also representing a group for which there is variation in what issues are important to their respective electorates (variation that does not exist among national-level politicians).

Finally, and related to the previous points, the Sweden Democrats’ electoral success provides a clear signal for politicians. The Sweden Democrats have made refugee immigration the center of their party platform and the centerpiece of their campaign. Until very recently, the Sweden Democrats have been the only competitive Swedish party in favor of restricting refugee immigration (cf. Aylott & Bolin, 2015; Dahlström & Esaiasson, 2013). In 2010, 93% of the MPs from the Sweden Democrats supported the idea of reducing the number of refugees in Sweden. The corresponding number for MPs from the other parties was, on average, 3% according to the 2010 Swedish Parliamentary Survey. Further, in the 2010 national election, 70% of the voters could associate the Sweden Democrats with at least one electoral issue, the average for the other parties was 57%. And in particular, 68% of the voters associated the Sweden Democrats with immigration and/or refugees. This gives the Sweden Democrats the strongest issue brand among all parties.⁴

In 2014, the issue of immigration again set the Sweden Democrats apart. In the election, political leaders from the other parties took strong stances against the reduction of refugee immigration. Prime Minister Fredrik Reinfeldt, leader of the large Conservative Party in Sweden, pleaded with Swedes to “be patient and to open their hearts and to see the people living under such harsh stress” (Ekot, 2014). Supporting continued refugee immigration became an important part of the campaign with the Sweden Democrats being the only party in favor of reducing immigration. Consequently, the Sweden Democrats’ electoral success provided a clear signal about what their supporters cared about. As a result, we have clear theoretical predictions about how individual politicians should respond to these results: If the Sweden Democrats did well in their home area, they should lobby leaders for reductions in immigration.

Description of the Survey

Our study is based on a survey of municipal politicians conducted in February 2015. The survey had great coverage across the country and we have responses from nearly 1,700 politicians, representing 269 of Sweden’s 290 municipalities. The solid line in Figure 1 gives the distribution of the change in the Sweden Democrats’ vote share in the local municipal elections between 2010 and 2014 (i.e., their vote share in 2014 minus their vote share in 2010) for the municipalities where politicians in our sample live, whereas the dashed line gives the distribution for all 290 municipalities. The distributions are nearly identical, showing that the electoral experience of our sample is reflective of the election shifts in Swedish municipal elections more generally. The figure also shows that, while the Sweden Democrats increased their share of the vote by an average of 5 percentage points across municipalities, there is important variation in their success. In some municipalities, the Sweden Democrats increased their vote share by as much as 18 percentage points, while in other places the Sweden Democrats did not increase their vote share (in only two municipalities did the Sweden Democrats do worse in 2014 than they did in 2010).

The 2015 survey was taken by politicians in the Panel of Politicians, which is an online university-based panel consisting of local-, regional-, and national-level politicians in Sweden and is conducted by the Laboratory of Opinion Research (LORE) at the University of Gothenburg. The Panel of Politicians recruits respondents via email addresses found on local, regional, and national parliament websites as well as via surveys of politicians⁵ where respondents at the end of the survey are asked to sign up for the panel. Panelists are also regularly asked to recommend other colleagues to participate in the panel. The response rate for the survey used in this article was about 50% of the panel, with nearly 1,700 politicians participating. We also use responses from a 2013 survey of panelists to get a prior estimate of respondents’ views on refugee immigration. In the 2013 survey, 1,300 of the politicians in our survey gave their view of immigration of refugees. We drop the politicians who are Sweden Democrats from our sample.

We asked politicians how much they lobbied both for and against limiting refugee immigration to track adaptation in both directions. The question was asked in the following way: “During the last 4 months, have you tried to influence the leadership of your party to take the following position?” “To reduce immigration of refugees,” “To not reduce immigration of refugees.” The answering alternatives were “No,” “Yes, a few times,” and “Yes, many times.”⁶ This allows us to test whether the Sweden Democrats’ increased electoral success predicts more lobbying to restrict refugee immigration and/or less lobbying for the opposite position.

We study lobbying on the issue of refugee immigration because this is how the debate on immigration is primarily framed in Sweden (Demker, 2012). Sweden, accepted more refugees per capita than all other Organization for Economic Co-Operation and Development (OECD) countries in 2013. On average, OECD countries welcomed 830 refugees per 1,000,000 citizens, while the number for Sweden was 5,700 refugees per 1,000,000 citizens (OECD, 2014, p. 26). Sweden alone accepted 22% of all refugees entering the European Union (Ruist, 2015). The most common country of birth of immigrants entering Sweden in 2014 was Syria, followed by Eritrea, Poland, and Somalia; and since the outbreak of the civil war in Syria, the Syrian population in Sweden has increased rapidly. In 2014, Sweden received more than 30,000 refugees from Syria and the number increased during 2015 (Statistics Sweden).

We study whether politicians report that they lobby their party’s leadership because this is how we expect politicians in a party-centered system, like Sweden, to pursue individual-level responsiveness. However, party loyalty does not mean that the politicians ignore their local public opinion. Instead, we expect politicians to internally lobby the party to move the party toward their constituents’ preferences.

We asked politicians about their lobbying during the “last 4 months” because the survey was conducted in February 2015, 5 months after the September 2014 municipal election and because we wanted to catch adaptation after the election rather than before the election. We chose to not explicitly mention the election itself because we wanted to minimize possible demand effects in how the politicians responded. About 60% of the sample reported not lobbying on either side of the issue. Another 15% of the officials reported lobbying party leaders to take a position in favor of reducing the number of refugees (12% reported doing so a few times and 3% doing so many times) and 27% reported lobbying party leaders against reducing the numbers of refugees (22% reported doing so a few times and 5% doing so many times). Two percent of the respondents in the sample reported lobbying party leaders on both sides of the issue. We include them in the main body of analysis, but we also checked to make sure the results held when excluding these respondents from the analysis. The results hold when dropping these respondents (see Online Appendix C).

The fact that there was a significant number of politicians on both sides of this issue is consistent with previous studies that show that the issue of refugee immigration is very polarized in Sweden (Hellström, Nilsson, & Stoltz, 2012). It also supports the idea that politicians are pressured to respond to the public and also safeguard the party’s position.⁷ The next section analyzes whether lobbying of party leadership correlates with the Sweden Democrats’ increased municipal electoral success.

Within-Party Adaptation in the Wake of Elections

If politicians react to the change in the Sweden Democrats’ vote share, we expect the officials from places where the Sweden Democrats increased their vote share the most to be more likely to lobby for (and less likely to lobby against) reducing the number of refugees. In other words, we expect a positive correlation between the change in the Sweden Democrats’ vote share in an official’s municipality and how much that official lobbied party leaders to publicly take a position for reducing the number of refugees. We also expect a negative correlation between the change in the Sweden Democrats’ vote share and how much the officials lobbied against reducing the number of refugee immigrants.⁸

Table 1 presents the simple, bivariate regressions testing the relationship between legislators’ internal lobbying and the local change in the Sweden Democrats’ vote share. Here, and throughout the article, we analyze the original 3-point scale (1 = no lobbying, 2 = lobbying a few times, 3 = lobbying many times) and also a binary indicator of whether the politician lobbied on the issue or not (0 if they did not lobby and 1 if they lobbied either a few times or many times). We present the additional recoded, binary outcome because the coefficients in those models are readily interpretable as the percentage point increase related to the change in the Sweden Democrats’ vote share. We use ordinary least squares (OLS) regression to estimate all of the models presented in the main body of the article. The results are also robust to using regression models designed for discrete data (see Online Appendix A). Throughout we also cluster the standard errors on the municipality to account for the fact that we have more than one respondent from many of the municipalities in the sample.

Table 1.

The Effect of Sweden Democrats’ Vote Share on Internal-Party Lobbying on the Issue of Restricting Refugee Immigration to Sweden.

Dependent Variable:	Lobbying FOR reduction		Lobbying AGAINST reduction
Dependent Variable:	3-point scale	Binary	3-point scale	Binary
Model:	OLS	OLS	OLS	OLS
Independent variables	(1)	(2)	(3)	(4)
Change in Sweden democrats’ Vote share (2014-2010)	0.019* (0.006)	0.013* (0.004)	0.000 (0.006)	0.001 (0.005)
Constant	1.068* (0.031)	0.074* (0.022)	1.326* (0.036)	0.269* (0.029)
R ²	.011	.008	.000	.000
Clusters (municipalities)	269	269	269	269
Observations	1,248	1,248	1,241	1,241

Respondents are municipal officials in Sweden. The 3-point scale outcomes (columns 1 and 3) are coded as follows: 1 = no lobbying, 2 = lobbying a few times, 3 = lobbying many times. The binary outcome (columns 2 and 4) is coded as 0 = no lobbying, 1 = lobbying (either a few times or many times). Standard errors clustered on municipality are given in parentheses. OLS = ordinary least squares.

p < .05.

The positive coefficients in columns 1 and 2 indicate that in areas where the Sweden Democrats increased their vote share more, politicians were more likely to lobby for reducing immigration of refugees. These coefficients are statistically significant and also substantively meaningful. For example, the results in column 2 indicate that an increase of 10 percentage points in the change in the Sweden Democrats’ vote share is correlated with the local official being 13 percentage points more likely to lobby for reducing the number of refugees. This represents a greater than one-to-one relationship and is strong evidence that local politicians respond to the election results by working within the party structure to get leaders to change their stance on the issues that successful parties are promoting.

There is no evidence for the flip side of this relationship; the success of the Sweden Democrats in the 2014 election is not related to whether politicians lobbied against reducing the number of refugees. The coefficients are not statistically significant, are close to zero, and do not always point in the predicted (negative) direction.

The Sweden Democrats’ electoral success may have done less to deter lobbying on the pro-immigration side of the debate because of a ceiling effect. It may simply be that the politicians lobbying for protecting Sweden’s pro-immigration policies are true believers who are committed to that position, no matter what signals voters send. Immigration policy was one of the issues where public policy in Sweden was the least congruent with voters’ preferences at the time (Holmberg, 2014). As a result, it may be that the politicians who could be moved from lobbying on the pro-immigration side had already moved. In other words, the politicians still lobbying to maintain Sweden’s immigration policy may be the very people who would have done so regardless of the election outcome.

Mitigating for Potential Omitted Variable Bias

While we are interested in the causal effect of election results on internal-party lobbying, our data are not experimental. As a result, drawing causal conclusions from the bivariate relationship is contingent upon assuming that changes in the Sweden Democrats’ electoral vote share is randomly distributed across municipalities. This is obviously not the case. This could be a problem for making inferences from the data if, for example, politicians who serve in places where the Sweden Democrats increased their vote share are systematically from certain parties or simply have different views about the immigration of refugees. If this is the case, then the positive correlation we have found may be spurious and attributable to other differences across these different types of municipalities.

In this section, we test whether the bivariate relationship holds up when we control for potentially relevant variables. This approach still relies on assumptions, but it relies on a less stringent assumption. In particular, this approach relies on the assumption that changes in the Sweden Democrat’s electoral vote share is independent (i.e., as if randomly distributed across municipalities) conditional on the control variables. In our analysis, we mitigate concerns about omitted variable bias by controlling for the Sweden Democrats’ vote share in the prior election (i.e., in 2010), the share of refugees in the municipality, and the politicians’ prior attitudes, party affiliation, and gender.

We control for the prior levels of electoral support for the Sweden Democrats as a way to control for relevant preexisting political differences. Similarly, we control for the share of refugees in the municipalities where the politicians are elected to be sure that the results are not driven by the actual refugee immigration situation facing the politicians. More specifically, we use statistics from the Swedish public service broadcaster’s research group “Pejl” (SVT, 2014). These statistics give the share of refugees per 1,000/inhabitants that arrived in each municipality during the period 2006 to 2013. Using this longer time frame allows us to control for how the long-term trends in immigration leading up to the election we study affect attitudes.

As for the individual-level factors, we are able to control for politician’s prior attitude because in 2013, a year before the election, politicians in the panel had been asked about their views on a proposal to accept fewer refugees: “What do you think of the proposal to accept fewer refugees into Sweden?” Politicians chose a response from a 5-point scale where increasing values indicate greater opposition to reducing the number of refugees (i.e., 1 = very good proposal,” 5 = very bad proposal). We also control for party because responsiveness at the aggregate party level (as measured via content of election manifestos) varies across different party types, with large parties being more responsive to voters’ policy priorities and government parties less responsive (Klüver & Spoon, 2014; see also Wagner & Meyer, 2014). Because we are looking at the effect of the Sweden Democrats on politicians from other parties, we exclude Sweden Democrats from the sample we analyze. Finally, prior evidence suggests that women are generally less negative toward refugee immigration in Sweden, and we therefore consider that in our model (Demker, 2012).

Table 2 shows that including the full set of control variables does not change the estimated coefficients sizes very much. The coefficient for the relationship between the change in the Sweden Democrats’ vote share and lobbying against reducing the number of refugees continues to be close to zero and statistically insignificant. The Sweden Democrats’ success did nothing to dissuade politicians who wanted to maintain the status quo from voicing their opinions to their leaders. At the same time, the results show that the Sweden Democrats’ vote share continues to be an important predictor of lobbying for reducing the number of refugees. The point estimate in Model 6 of Table 2 suggests that when the Sweden Democrats get 10 percentage points more of the vote, the local politicians are 9 percentage points more likely to lobby their party leaders to take a more restrictive stance on the number of refugees that should be accepted in Sweden.

Table 2.

The Effect of Sweden Democrats’ Vote Share on Internal-Party Lobbying—Full Model.

Dependent variable:	Lobbying FOR reduction		Lobbying AGAINST reduction
Dependent variable:	3-point scale	Binary	3-point scale	Binary
Model:	OLS	OLS	OLS	OLS
Independent variables	(5)	(6)	(7)	(8)
Change in Sweden democrats’ Vote share (2014-2010)	0.014* (0.006)	0.009* (0.004)	0.005 (0.007)	0.004 (0.005)
Control variables
Sweden Democrats’ lagged vote share (2010)	0.005 (0.004)	0.004 (0.003)	−0.004 (0.005)	−0.001 (0.004)
Female official	−0.007 (0.021)	−0.006 (0.017)	0.002 (0.034)	−0.008 (0.026)
Refugee share	0.000 (0.001)	0.000 (0.001)	0.000 (0.001)	0.000 (0.001)
Lagged attitude on immigration policy (2013)	−0.104* (0.012)	−0.087* (0.010)	0.106* (0.015)	0.078* (0.012)
Official’s individual partisanship (omitted, baseline group = Left Party)
Center Party	0.061 (0.039)	0.062 (0.033)	0.008 (0.073)	0.042 (0.058)
Social Democrat	0.012 (0.017)	0.013 (0.015)	0.169* (0.066)	0.148* (0.048)
Liberal Party	0.050 (0.037)	0.047 (0.032)	0.119 (0.079)	0.111 (0.058)
Conservative	0.269* (0.041)	0.227* (0.034)	0.031 (0.072)	0.045 (0.056)
Christian Democrat	0.019 (0.039)	0.022 (0.035)	0.101 (0.088)	0.127 (0.070)
Green Party	−0.031 (0.020)	−0.024 (0.018)	0.010 (0.087)	0.011 (0.063)
Feminist initiative			0.224 (0.434)	0.294 (0.416)
Constant	1.376* (0.070)	0.324* (0.056)	0.838* (0.102)	−0.118 (0.079)
R ²	.229	.234	.074	.067
Clusters (municipalities)	269	269	269	269
Observations	1,248	1,248	1,243	1,243

Respondents are municipal officials in Sweden. All models are estimated via OLS. The 3-point scale outcomes (columns 5 and 7) are coded as follows: 1 = no lobbying, 2 = lobbying a few times, 3 = lobbying many times. The binary outcome (columns 6 and 8) is coded as 0 = no lobbying, 1 = lobbying (either a few times or many times). Standard errors clustered on municipality are given in parentheses. OLS = ordinary least squares.

p < .05.

The coefficients for the control variables are also worth noting. First, we find that legislators’ attitudes on the issue (which were measured 2 years prior to the question on lobbying) systematically predicts levels of lobbying. Consistent with previous research, the results in columns 5 and 6 show that politicians’ own attitudes predicted how much they lobbied leaders on this issue (Burden, 2007). Politicians who thought the proposal to reduce refugee immigration was a bad proposal were less likely to lobby for reducing the number of refugees. The magnitude of this effect is quite large. The results from column 6 suggest that those who thought the proposal was a very bad proposal were about 35 percentage points less likely to lobby leaders to reduce refugee immigration than officials who thought it was a very good proposal.⁹

Most importantly, even controlling for their prior attitudes, we see that the Sweden Democrats’ vote share has a substantial effect on the likelihood of lobbying on this policy issue. Recall that a 10 percentage point increase in the Sweden Democrats’ vote share leads politicians to be about 9 percentage points more likely to lobby party leaders. By comparison, there is a 35 percentage point difference between those who thought that reducing immigration was a very bad proposal and those who thought it was a very good proposal. Thus, a 10 percentage point shift in the party’s electoral success has an effect that is close to about one fourth the size of completely changing one’s mind on the issue and going from being strongly against reductions to being strongly for reductions. Elections are not the most powerful predictor of lobbying, but they are a meaningful one.

Second, we find that women are not more likely to lobby on either side of this issue. While previous studies have found that women are generally less negative toward refugee immigration in Sweden (Demker, 2012), we do not find that gender predicts lobbying on either side of this issue. One reason we may not have found a gender difference is that our model controls for the respondent’s attitude. It may be that women lobby more because they are more supportive of immigration (as they are in our sample), but we are directly capturing that effect by controlling for their prior attitude on the issue. Our results simply show that gender does not predict the likelihood that a politician lobbies party leaders on this issue above and beyond the effect of one’s own attitude.

Third, we find that the Sweden Democrat’s previous vote share and the share of refugees are not statistically significant predictors of whether legislators lobby on the issue. Finally, we find some differences between the parties. The Left Party and the Green Party were at the time advocates for even more generous policies toward refugees (Dahlström & Esaiasson, 2013). Accordingly, we find that representatives from these two parties are less likely than other parties to lobby for reductions in refugee immigration when holding the other factors constant.¹⁰

Did Politicians Change Their Own Minds on the Issue?

Our study is motivated by a desire to test whether politicians learn about constituents’ preferences from elections and then work within the party structure to adapt and represent those preferences. This representation story often implicitly assumes that politicians adapt to constituents’ preferences without actually updating their own preferences (Grossback, Peterson, & Stimson, 2005). However, that is not the only model of representation. Alternatively, representation may be achieved because the elections have a mediating effect on the politicians’ own attitudes (Miller & Stokes, 1963).

We test for the plausibility that politicians’ own attitudes are mediating the effect of the Sweden Democrats’ election success by running models where the proposed mediator—politicians’ postelection attitude about immigration of refugees—is the dependent variable (Bullock, Green, & Ha, 2010).¹¹ The results, which are presented in Table 3, show that the Sweden Democrats’ vote share has no statistically significant effect on politicians’ 2015 attitude about immigration of refugees. Elections are thus related to politicians’ advocacy behavior within their parties but not their own opinions. This is further evidence that information about voter preferences influences politicians; it can even make them act as delegates who advocate for positions that they do not personally hold.

Table 3.

Effect of Sweden Democrat Vote Share on 2015 (Postelection) Attitude.

Dependent variable:	2015 refugee policy attitude
Model:	OLS
Independent variables	(9)
Change in Sweden Democrats’ Vote Share (2014-2010)	−0.003 (0.010)
Control variables
Sweden Democrats’ lagged vote share (2010)	−0.002 (0.009)
Female official	0.235* (0.055)
Refugee share	0.002 (0.002)
Lagged attitude on immigration policy (2013)	0.578* (0.032)
Official’s individual partisanship (omitted, baseline group = Left Party)
Center Party	−0.469* (0.115)
Social Democrat	−0.241* (0.080)
Liberal Party	−0.622* (0.113)
Conservative	−1.060* (0.119)
Christian Democrat	−0.502* (0.126)
Green Party	−0.056 (0.106)
Feminist Initiative	0.517 (0.344)
Constant	2.036* (0.196)
R ²	.528
Clusters (municipalities)	269
Observations	1,246

Respondents are municipal officials in Sweden. The dependent variable is the official’s 2015 (postelection) response to the question “What do you think of the proposal to accept fewer refugees into Sweden?” Answer options: 1 = very good proposal, 2 = fairly good proposal, 3 = neither good nor bad proposal, 4 = fairly bad proposal, 5 = very bad proposal. Standard errors clustered on municipality are given in parentheses. OLS = ordinary least squares.

p < .05.

Discussion

We investigated how individual politicians respond to signals sent by voters in elections by studying how they work to change their party from within. We make use of the Sweden Democrats’ increased electoral success in the 2014 Swedish municipal elections and find that municipal politicians serving in the areas where the Sweden Democrats increased their vote share by a greater amount are more likely to lobby their party leaders to adopt a more restrictive position on the immigration of refugees. We find that these results hold even when we control for potential omitted variable bias in the form of politicians’ own opinions, their party affiliation, the share of refugees in the municipality, and the Sweden Democrat’s prior electoral success in the area.

Contingent upon the assumption that the Sweden Democrat’s electoral vote share is independent conditional on these control variables, our analysis shows that politicians’ response to the election results was meaningful. With a 10 percentage point increase in the Sweden Democrats’ electoral success in a municipality, the local officials in that municipality were 9 percentage points more likely to lobby their leaders to take a public stance on reducing the number of refugees. While that is a large number of votes, it is close to the increase in vote share that the Sweden Democrats have recently enjoyed. In 2006, the Sweden Democrats only received 2.9% of the votes for the parliament. In 2014, just 8 years later, they were receiving 12.9 percentage points. Over an 8-year period, the Sweden Democrats have seen a 10 percentage point increase in their vote share in national elections. These results suggest that the rank-and-file politicians in the other parties are likely to be about 9 percentage points more likely to lobby their party leaders to reduce refugee immigration than they would have been had the Sweden Democrats not achieved these levels of electoral success.

Of course, the fact that we study one country, one issue, and one niche party’s success affects the generalizability of our results. Because the Swedish electoral system is highly proportional, all votes are “useful” in the sense that even small parties have a high chance of entering the parliament. In this way, politicians have incentives to be responsive to changes in vote share. Furthermore, we study a highly salient and polarizing issue in this system—refugee immigration in the 2014 election. Prior evidence indicates that when voters are more divided on an issue, parties might be more responsive (Spoon & Klüver, 2015; see also Abou-Chadi, 2014). On the other hand, the immigration issue was one that divided the electorate and the elite at the time in Sweden (Holmberg, 2014) with all of the other parties taking a stance against restricting immigration. This should have functioned as a dampening effect on the individual politicians’ responsiveness. Furthermore, Swedish politicians are particularly loyal party representatives and are likely to often follow the will of the party before the will of their constituents (Erlingsson et al., 2016), also making it potentially harder to find an effect in this context. Taken together, our work provides evidence of a previously unseen phenomenon that is likely to exist in all contexts where parties, candidates, and voters interact. But it remains to be further clarified how this phenomenon varies between political contexts.

Another interesting question for further study is whether other signals about citizens’ preferences, such as protests, are equally effective at causing politicians to lobby party leaders to change policy positions. On one hand, we might expect elections to be more effective because politicians should be particularly concerned about the people who show up and vote (i.e., the signal they receive from elections). On the other hand, elections are time-bound. Some of these other signals, such as protests and opinion polls, are often more recent and may provide a better signal of where voters are today than a not-so-recent election. Further empirical work could help determine the relative strength of these different signals.

Still, our results already have the potential to change how we think about representation and responsiveness in strong-party systems. In particular, we show that individual politicians in strong-party systems advocate on behalf of their constituents by working within the party system (see also Öhberg & Naurin, 2015). Because party leaders in strong-party systems have numerous carrots and sticks they can use on rank-and-file members (Carey, 2007, 2009; Strøm, 1997), a considerable part of theorizing about parties views them as monolithic structures. However, while parties might be seen as coherent actors from the outside, important dynamics occur within the party. Our study shows that politicians clearly engage in lobbying on the important issues of the day. While some of this lobbying relates to members’ own preferences on the issues, it also reflects voters’ positions. We have shown that individual politicians respond to the election results by changing the positions they advocate for. If we do not account for these actions, we will underestimate the level of representation in party systems.

We regard our findings as complimentary to the existing literature on how parties respond to changes in other parties’ electoral fortunes in that we highlight an individual-level mechanism that helps explain how parties respond to election results to shift policy over time (Klüver & Spoon, 2014; Spoon et al., 2014; cf. Meguid, 2005, 2008). Hypotheses about “external stimuli” (Harmel & Janda, 1994) and “lagged effects of responsiveness,” where parties need time to adjust their issue priorities (Adams & Somer-Topcu, 2009; Spoon & Klüver, 2015), are in line with our findings. So are hypotheses about the effect of niche party success on mainstream parties’ policy shifts (Bale et al., 2010; Spoon et al., 2014). The central difference between our results and past work lies in our investigation of the individual-level reactions of politicians to voter signals. Our results provide the micro-foundations to explain why parties, at the aggregate level, respond to electoral shifts.

Understanding legislators’ lobbying behavior also provides a rationale for why smaller niche parties might run for office, even if they never become part of the ruling coalition. In particular, our results show that the electoral success of parties focusing on a niche issue can lead the politicians in other, mainstream parties to be more willing to advocate internally for those positions. While serving in the ruling coalition is one way to achieve one’s policy goals, parties can also achieve policy success by getting other parties to implement those policy goals. This approach is incentive compatible because although niche parties are punished for shifting positions, mainstream parties are rewarded for such moves (Adams et al., 2006, see also Hobolt & de Vries, 2015). Niche parties thus have incentives to run for office as a way of affecting the political positions of officials in larger, more mainstream parties.

Further, our results add to the discussion of how the success of anti-immigration parties affects mainstream parties. The high and rising pressure on European democratic systems to accept immigrants from the Middle East and Africa makes this an increasingly important issue. While there are studies on how policy shifts in parties’ official programs (see Bale et al., 2010), our study probably speaks to what happens right before such shifts. If enough politicians advocate internally for change, it is possible that parties change suddenly once they reach a tipping point. Because the work that politicians put into changing their party from within is not always visible, these changes might catch outside observers by surprise (Kuran, 1991). Our results provide an underlying mechanism that explains how policy changes on this issue, as well as others, and how it has the potential to be sudden and surprising. Future studies should pay more attention to how party leadership reacts to inside lobbying from their own representatives who are trying to be responsive to the electorate.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: Dan Butler thanks the Wedeinbaum Center at WUSTL for financial support. Elin Naurin and Patrik Öhberg are grateful to FORTE for providing valuable research funding.

Notes

Author Biographies

Daniel M. Butler is associate professor of political science at Washington University in St. Louis. His work focuses on representation and the behavior of elites in the United States and Europe.

Elin Naurin is associate professor at the department of political science, University of Gothenburg. Her work includes studies of responsiveness, parties’ election pledge fulfilment and voters’ perception of electoral mandates.

Patrik Öhberg is assistant professor at the department of political science, Univeristy of Gothenburg. He is the initator and principle investigator of the Panel of Politicians. His work gives focus to representation and behaviour of political elites.

References

Abou-Chadi

(2014). Niche party success and mainstream party policy shifts—How green and radical right parties differ in their impact. British Journal of Political Science, 46, 417-436.

Adams

(2012). The causes and the electoral consequences of party policy shifts in multiparty elections: Theoretical results and empirical evidence. Annual Review of Political Science, 15, 401-419.

Adams

Clark

Ezrow

Glasgow

(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, 513-529.

Adams

Haupt

A. B.

Stoll

(2009). What moves parties? The role of public opinion and global economic conditions in Western Europe. Comparative Political Studies, 42, 611-639.

Adams

Somer-Topcu

(2009). Policy adjustment by parties in response to rival parties’ policy shifts: Spatial theory and the dynamics of party competition in twenty-five post-war democracies. British Journal of Political Science, 39, 825-846.

American Political Science Association Committee on Political Parties. (1950). Toward a more responsible two-party system. American Political Science Review, 44(Suppl.) .

Arceneaux

Johnson

Lindstädt

Wielen

R. J. V.

(2015). The influence of news media on political elites: Investigating strategic responsiveness in Congress. American Journal of Political Science, 60, 5-29.

Aylott

Bolin

(2015). Polarising pluralism: The Swedish parliamentary election of September 2014. West European Politics, 38, 730-740.

Bafumi

Herron

M. C.

(2010). Leapfrog representation and extremism: A study of American voters and their members in Congress. American Political Science Review, 104, 519-542.

10.

Bale

Green-Pedersen

Krouwel

Luther

K. R.

Sitter

(2010). If you can’t beat them, join them? Explaining social democratic responses to the challenge from the populist radical right in Western. Political Studies, 58, 410-426.

11.

Belanger

Meguid

B. M.

(2008). Issue salience, issue ownership, and issue-based vote choice. Electoral Studies, 27, 477-491.

12.

Berg

Oscarsson

H. H.

(Eds.). (2012). Omstritt omval. Gothenburg [Disputed reelection], Sweden: SOM-institutet, Göteborgs universitet.

13.

Bowler

(2010). Private members’ bills in the UK parliament: Is there an “electoral connection”? Journal of Legislative Studies, 16, 476-494.

14.

Broockman

D. E.

(2016). Approaches to studying policy representation. Legislative Studies Quarterly, 41, 181-215.

15.

Broockman

D. E.

Butler

D. M.

(2015). The Causal Effects of Elite Position-Taking on Voter Attitudes: Field Experiments with Elite Communication. American Journal of Political Science. Advance online publication. doi:10.1111/ajps.12243

16.

Budge

(1994). A new theory of party competition. British Journal of Political Science, 24, 443-467.

17.

Budge

Ezrow

McDonald

M. D.

(2010). Ideology, party factionalism and policy change: An integrated dynamic theory. British Journal of Political Science, 40, 781-804.

18.

Bullock

J. G.

Green

D. P.

S. E.

(2010). Yes, but what’s the mechanism? (Don’t expect an easy answer). Journal of Personality and Social Psychology, 98, 550-558.

19.

Burden

B. C.

(2007). Personal roots of representation. Princeton, NJ: Princeton University Press.

20.

Butler

D. M.

Nickerson

D. W.

(2011). Can learning constituency opinion affect how legislators vote? Results from a field experiment. Quarterly Journal of Political Science, 6, 55-83.

21.

Carey

J. M.

(1996). Parties, incentives, and term limits in Costa Rica. In Grofman

(Ed.), Term limits: Public choice perspectives (pp. 321-346). Boston, MA: Kluwer Academic.

22.

Carey

J. M.

(2007). Political institutions, competing principals, and party unity in legislative voting. American Journal of Political Science, 51, 92-107.

23.

Carey

J. M.

(2009). Legislative voting and accountability. Cambridge, MA: Cambridge University Press.

24.

Carey

J. M.

Shugart

M. S.

(1995). Incentives to cultivate a personal vote: A rank ordering of electoral formulas. Electoral Studies, 14, 417-439.

25.

Dahlström

Esaiasson

(2013). The immigration issue and anti-immigrant party success in Sweden 1970-2006: A deviant case analysis. Party Politics, 19, 343-364.

26.

Dancygier

Lindgren

K.-O.

Oskarsson

Vernby

(2015). Why are immigrants underrepresented in politics? Evidence from Sweden. American Political Science Review, 109, 703-724.

27.

Demker

(2012). Scandinavian right-wing parties. In Mammone

Godin

Jenkins

(Eds.), Mapping the extreme right in contemporary Europe (pp. 239-253). London, England: Routledge.

28.

Downs

(1957). An economic theory of democracy. New York, NY: Harper.

29.

Dudzinska

Poyet

Costa

Wessels

(2014). Representational roles. In Deschouwer

Depauw

(Eds.), Representing the people: A survey among members of statewide and substate parliaments (pp. 19-38). Oxford, UK: Oxford University Press.

30.

Ekot. (2014). http://sverigesradio.se/sida/artikel.aspx?programid=83&artikel=5939422

31.

Erlingsson

Kölln

A.-K.

Öhberg

(2016). Swedish party organisations. In Pierre

(Ed.), Oxford handbook of Swedish politics (pp. 169-187). Oxford, UK: Oxford University Press.

32.

Ezrow

Hellwig

(2014). Responding to voters or responding to markets? Political parties and public opinion in an era of globalization. International Studies Quarterly, 58, 816-827.

33.

Fagerholm

(2015). Why do political parties change their policy positions? A review. Political Studies Review, 14, 501-511.

34.

Gauja

(2013). Policy development and intra-party democracy. In Cross

W. P.

Katz

R. S.

(Eds.), The challenges of intra-party democracy (pp. 116-135). Oxford, UK: Oxford University Press.

35.

Golder

Stramski

(2010). Ideological congruence and electoral institutions. American Journal of Political Science, 54, 90-106.

36.

Green-Pedersen

(2007). The growing importance of issue competition: The changing nature of party competition in Western Europe. Political Studies, 55, 607-628.

37.

Grossback

L. J.

Peterson

D. A. M.

Stimson

J. A.

(2005). Comparing competing theories on the causes of mandate perceptions. American Journal of Political Science, 49, 406-419.

38.

Harmel

Janda

(1994). An integrated theory of party goals and party change. Journal of Theoretical Politics, 6, 259-287.

39.

Hazan

R. Y.

Rahat

(2010). Democracy within parties: Candidate selection methods and their political consequences. Oxford, UK: Oxford University Press.

40.

Hellström

Nilsson

Stoltz

(2012). Nationalism vs. nationalism: The challenge of the Sweden Democrats in the Swedish public debate. Government & Opposition, 47, 186-205.

41.

Hobolt

S. B.

de Vries

C. E.

(2015). Issue entrepreneurship and multiparty competition. Comparative Political Studies, 48, 1159-1185.

42.

Holmberg

(2014). Representerar riksdagen svenska folket? [Does the Riksdag represent the Swedish people?] In Karlsson

Gilljam

(Eds.), Svenska politiker. Om de folkvalda i riksdag, landsting och kommun (pp. 93-116). Stockholm, Sweden: Santerus förlag.

43.

Jensen

T. K.

(2000). Party cohesion. In Esaiasson

Heidar

(Eds.), Beyond Westminster and Congress: The Nordic experience (pp. 210-236). Columbus: Ohio State University Press.

44.

Klingemann

H.-D.

Hofferbert

R. I.

Budge

(1994). Parties, policies, and democracy. Boulder, CO: Westview Press.

45.

Klüver

Spoon

J.-J.

(2014). Who responds? Voters, parties and issue attention. British Journal of Political Science, 46, 633-654.

46.

Kousser

Lewis

J. B.

Masket

S. E.

(2007). Ideological adaptation? The survival instinct of threatened legislators. The Journal of Politics, 69, 828-843.

47.

Kuran

(1991). Now out of never: The element of surprise in the East European revolution of 1989. World Politics, 44, 7-48.

48.

Lijphart

(1984). Democracies: Patterns of majoritarian and consensus government in twenty-one countries. New Haven, CT: Yale University Press.

49.

Lindstädt

Vander Wielen

R. J.

(2014). Dynamic elite partisanship: Party loyalty and agenda setting in the US house. British Journal of Political Science, 44, 741-772.

50.

Mayhew

D. R.

(1974). Yale Studies in Political Science: Congress: The electoral connection (Vol. 26). New Haven, CT: Yale University Press.

51.

McDonald

Budge

(2005). Elections, parties, democracy: Conferring the median mandate. Oxford, UK: Oxford University Press.

52.

Meguid

B. M.

(2005). Competition between unequals: The role of mainstream party strategy in niche party success. American Political Science Review, 99, 347-359.

53.

Meguid

B. M.

(2008). Party competition between unequals: Strategies and electoral fortunes in Western Europe. Cambridge, UK: Cambridge University Press.

54.

Miller

W. E.

Stokes

D. E.

(1963). Constituency influence in Congress. American Political Science Review, 57, 45-56.

55.

Monroe

A. D.

(1979). Consistency between constituency preferences and national policy decisions. American Politics Quarterly, 12, 3-19.

56.

Nord

L. W.

(2007). The Swedish model becomes less Swedish. In Negrine

Mancini

Holtz-Bacha

Papathanassopoulos

(Eds.), The professionalisation of political communication (pp. 81-95). Bristol, UK: Intellect.

57.

Öhberg

Naurin

(2015). Party-constrained policy responsiveness: A survey experiment on politicians’ response to citizen-initiated contact. British Journal of Political Science, 46, 785-797.

58.

Organization for Economic Co-Operation and Development. (2014). International migration outlook 2014. Paris, France: Author.

59.

Powell

G. B.

Jr. (2000). Elections as instruments of democracy: Majoritarian and proportional visions. New Haven, CT: Yale University Press.

60.

Preece

J. R.

(2014). How the party can win in personal vote systems: The “selectoral connection” and legislative voting in Lithuania. Legislative Studies Quarterly, 39, 147-167.

61.

Proksch

S.-O.

Slapin

J. B.

(2012). Institutional foundations of legislative speech. American Journal of Political Science, 56, 520-537.

62.

Rahat

Hazan

R. Y.

(2001). Candidate selection methods: An analytical framework. Party Politics, 7, 297-322.

63.

Ruist

(2015, February). Refugee immigration and public finances in Sweden (Working Papers in Economics No. 613). Göteborg, Sweden: Göteborgs universitet.

64.

Somer-Topcu

(2009). Timely decisions: The effects of past national elections on party policy change. Journal of Politics, 71, 238-248.

65.

Somer-Topcu

Zar

M. E.

(2014). European parliamentary elections and national party policy change. Comparative Political Studies, 47, 878-902.

66.

Spoon

J.-J.

Hobolt

S. B.

de Vries

C. E.

(2014). Going green: Explaining issue competition on the environment. European Journal of Political Research, 53, 363-380.

67.

Spoon

J.-J.

Klüver

(2015). Voter polarisation and party responsiveness: Why parties emphasise divided issues, but remain silent on unified issues. European Journal of Political Research, 54, 343-362.

68.

Statistics Sweden. (2011). Eight parties election 2010 (General elections, election study). Retrieved from http://www.scb.se/statistik/_publikationer/ME0106_2010A01_BR_ME05BR1101.pdf

69.

Stimson

J. A.

MacKuen

M. B.

Erikson

R. S.

(1995). Dynamic representation. American Political Science Review, 89, 543-565.

70.

Stratmann

(2000). Congressional voting over legislative careers: Shifting positions and changing constraints. American Political Science Review, 94, 665-676.

71.

Strøm

(1997). Rules, reasons and routines: Legislative roles in parliamentary democracies. The Journal of Legislative Studies, 3, 155-174.

72.

Sveriges Television (STV) (2014). The Swedish public broadcaster’s “Pejl.” Retrieved from http://pejl.svt.se/val2014/flyktingmottagande-data/?page=14&offset=260

73.

van de Wardt

(2015). Desperate needs, desperate deeds: Why mainstream parties respond to the issues of niche parties. Western European Politics, 38, 93-122.

74.

Wagner

Meyer

T. M.

(2014). Which issues do parties emphasise? Salience strategies and party organisation in multiparty systems. West European Politics, 37, 1019-1045.

75.

Wlezien

Soroka

(2007). The relationship between public opinion and public policy. In Dalton

Klingemann

H.-D.

(Eds.), Oxford handbook of political behavior (pp. 799-817). Oxford, UK: Oxford University Press.

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

0.02 MB