Responsive Governance in the European Union

Transcription

1 Responsive Governance in the European Union Christina Schneider University of California, San Diego Abstract Intergovernmental organizations play a vital role in countries around the world, but little is known about the extent to which government behavior in these IOs is responsive to their national constituents. This paper analyzes whether European governments are responsive to their national electorates when they cooperate in the European Union. I argue that the politicization of European affairs pressures EU governments to signal responsiveness in European cooperation. I test the empirical implications of my theory using data on the bargaining behavior and negotiation success of 28 EU members in European legislative negotiations, and original data from a survey experiment in Germany. The findings suggest that EU governments are more likely to defend positions that favor their domestic constituents, and they will bargain harder to achieve successful negotiation outcomes, especially prior to national elections. Voters respond favorably to these signals of responsiveness. They prefer politicians who take their favored positions on policy issues, defend these positions, and who shift the final outcomes closer to the favored position. The results demonstrate that EU governments are responsive to their citizens views even when electoral accountability is low. Keywords: European Union, national elections, responsive government, international cooperation, democratic governance cjschneider@ucsd.edu. I have many friends and colleagues to thank for their helpful input at various stages of the project. I am grateful to Marisa Abrajano, Claire Adida, Michael Bechtel, Christophe Crombez, Peter Gourevitch, Emilie Hafner-Burton, Steph Haggard, Seth Hill, Simon Hug, David Lake, Dirk Leuffen, Helen Milner, Mark Pollack, Ken Scheve, Branislav Slantchev, Kaare Strom and the participants of the the HALBI workshop, and the political science workshop at the University of Konstanz for their very helpful comments. I am grateful to Abigail Vaughn for excellent research assistance. I gratefully acknowledge financial support from the Lifelong Learning Programme of the European Union and the UCSD Academic Senate.

2 Intergovernmental organizations (IOs) have proliferated since the end of World War II, and they play a vital role in countries around the world. Governments increasingly delegate decision-making powers to these organizations in areas that directly affect their sovereign autonomy, and it is commonly believed that IOs membership offers important benefits. As IOs have increasingly involved themselves in the domestic affairs of their member states, so has the criticism that decisions are taken out of voters hands and passed onto distant and sometimes even unelected political elites. Many commentators assert that IOs are elitist and technocratic. Decisions that are taken in IOs tend to be undemocratic and illegitimate because governments and bureaucrats are not accountable to domestic publics. They believe that international organizations suffer from a democratic deficit. 1 This legitimacy crisis has afflicted many international integration projects around the world, including the European Union (EU), Mercosur, North Atlantic Free Trade Association (NAFTA), the World Trade Organization (WTO), and even the United Nations (UN). 2 As more and more policies are decided in these international organizations, the pressure to democratize them has increased. The U.S. State Department, for example, considered a democratization of IOs as one of its main goals to improve their legitimacy and viability already in the early 2000s. 3 Nowhere has this been more salient than in the EU, where several dramatic setbacks in the past decade have further fueled the EU s legitimacy crisis. Starting with the Greek debt crisis in early 2010, which brought the Eurozone to its almost collapse several times, the situation has not gotten any better for one of the most ambitious projects of regional integration in the world. Still reeling from the economic and political turmoil of the European financial crisis, the EU was faced with its most significant external security crisis since the end of the Cold War when Russia annexed the Ukrainian peninsula of Crimea in spring of 2014 leading to a civil war that has caused the loss of thousands of lives so far. Doubts about the EU s ability to cope with external security questions culminated during 2015 when EU member countries were unable to respond collectively to the inflow of an unprecedented number of refugees into Europe and to thwart two major terrorists attacks in Paris and Brussels. And just when one thought that the 1 For a discussion of the democratic deficit in international organizations see, among many others, Dahl (1999); Nye (2001); Rohrschneider (2002); Scharpf (2006); Crombez (2003); Hix and Follesdal (2006); Moravcsik (2008). 2 See, for example, Anderson (1999); Malamud (2008); Zweifel (2006); Joseph (2011); Zaum (2013); Dellmuth and Tallberg (41). 3 See, for example, statements by Kim R. Holmes, Assistant Secretary for International Organizations Affairs in 2003 and p/io/rls/rm/2003/26949.htm and htm, last accessed: November

3 situation could not get any worse, the British population unexpectedly decided to leave the EU in a popular referendum in June 2016 causing major political and economic instability in the region. The Council of the European Union, which is the EU s main intergovernmental decision-making body and consists of the ministers from the elected governments of the member states, finds itself at the center of the EU s legitimacy crisis. In 2013, only 33% of Europeans trusted the Council, while over 44% of Europeans distrusted it. 4 Europeans feel that in European affairs their voice is not listened to by their governments. Governments are not responsive to their needs when they decide (usually behind closed doors) over policies in the EU. The argument that voters have lost influence over their own governments is superficially appealing (not the least because voters seem to believe it), but is this, in fact, the case? Despite its ever deeper and wider penetration into domestic policy, little is known about the degree to which EU governments are responsive to the views of their citizens when they cooperate at the EU-level. This paper presents an account of EU governments responsiveness to the opinions of their national constituents. In an effort to bridge the gap between previous findings about the politicization of European affairs at the national level and responsive governance in the EU, the paper embeds models of national electoral politics into models of intergovernmental cooperation using national elections as one important linkage point, and analyzes when and how governments appear responsive to their citizens before national elections. I argue that governments operate under uncertainty about whether European issues will become electorally relevant domestically. Consequently, they want to signal to their electorate that they take positions that are in the constituencies interest, and that they are willing to defend these positions throughout the negotiation process. They also want to claim credit for negotiation outcomes that are favorable to their constituents. EU governments are responsive to their domestic constituencies particularly before national elections, when their accountability is likely the greatest. To test my theoretical argument, I analyze data on the responsive governance of 28 EU governments in European legislative negotiations between 1998 and The findings suggest that EU members bargaining strategies and bargaining success crucially depend on the electoral cycles at the domestic level. Governments that face elections are less likely to move from their initial bargaining position and they are more likely to achieve bargaining outcomes that are close to their ideal positions. To further probe the 4 Data from the Interactive Eurobarometer. The legitimacy crisis afflicts other European institutions as well, including the European Parliament and the European Commission. For a discussion of the democratic deficit in these institutions see, for example, Hix (2008). 2

4 demand side of my theoretical argument, I present the results of a conjoint experiment that I conducted in a survey of over 2,500 Germans in the fall of I analyze how publics respond to different signals of responsiveness; a central assumption of my theory and one that had not received any empirical scrutiny before. I asked respondents to evaluate various politicians who differ on a set of responsiveness signals that correspond to the dimensions of theoretical interest as well as other characteristics of the politicians that may have an impact on their government approval. I find that voters are more likely to vote for politicians who represent their favored policy position, defend this position throughout the negotiations, and those who are successful in achieving their preferred outcome. Similarly, voters blame governments for representing unfavored positions, and for their inability to achieve favored negotiation outcomes. The findings provide insights into the question of democratic governance in international organizations. Whereas the research on the democratic responsiveness of governments has focused on government behavior at the national or subnational level, 5 the increasing role of IOs has increased pressures for more democratic governance at the international level. The minimum benchmark for representative democracies is that the views of citizens should influence government decisions (Achen, 1978). The findings in this paper demonstrate that EU governments are responsive to their domestic constituents in European negotiations, quite similar to their responsiveness at the national level, even though voters do not always hold them accountable for these actions. The Politicization of European Cooperation Are governments responsive to their domestic constituents when they cooperate at the EU level? This is not an idle question. Similar to many other international organizations, the European Union was an elitist, strategically depoliticized, operation from its inception (Schmidt, 2006; Mair, 2007). As envisioned by one of its principal architects, Jean Monnet, none of the major institutions of the European Coal and Steel Community the European Commission, the Council, and the European Court of Justice were created with direct democratic mandates. The idea was to speed up the integration process by minimizing the politicization of negotiations at the European level (Moravcsik, 1994). Jean Monnet even used an information obstruction policy, and asked news agencies not to cover the European Coal and Steel Community because he was worried that increasing public salience 5 See, for example, Stimson, MacKuen and Erikson (1995); Lax and Phillips (2009); Tausanovitch and Warshaw (2014). 3

5 would derail further integration (Atikcan, 2015). As a direct consequence, the EU was not politicized domestically, which also meant that voters preferences over European integration did not influence their vote choice in national elections. Voters were uninterested and uninformed, and European integration seemed to proceed in the shadow of a diffuse feeling of approval, or a permissive consensus (Lindberg and Scheingold, 1970). As a result, domestic electoral considerations played no role in European-level negotiations. Or so it seemed. Many observers note that the sleeping giant woke up (Van der Eijk and Franklin, 2004; Van der Brug, Van der Eijk and Franklin, 2007). 6 Issues of European integration gained in political salience with the Single European Act in 1986, because it expanded the range of European competencies to include policy areas like environmental protection, safety at work, and consumer protection (Scharpf, 2003, 6). The Maastricht Treaty of 1992 further accelerated the politicization of European integration (Hooghe and Marks, 2009). There is mounting evidence that attitudes toward European integration do affect vote choices, particularly when the issue (i) is mobilized by Euroskeptic issue entrepreneurs (de Vries, 2007, 2010), or (ii) becomes politicized at the domestic level through public debates between the government, opposition, the media, and other relevant groups (Franklin and Wlezien, 1997; Gable, 2000). The proliferation of referenda on EU matters, the rise of Euroskeptic parties, and the politicization of EU issues in national and European elections, has moved public opinion toward a constraining dissensus (Hooghe and Marks, 2009). The politicization of European cooperation at the national level should give ample incentives to governments to signal responsiveness to their voters at the European level. But even though we have made great strides toward understanding both the domestic politics of the EU as well as the intergovernmental politics of the EU, 7 we lack an understanding of how domestic politics affect the way European leaders negotiate and cooperate in the EU. That is, we do not know whether they are democratically responsive to the policy demands of their citizen. One notable exception is the work by Hagemann, Hobolt and Wratwil (2016). The authors analyze EU governments responsiveness, but they focus on analyzing how pro- or anti- European sentiments in the population can affect EU governments willingness to dissent in the voting process. 8 Whereas the influence of diffuse regime support on government behavior in the EU is without doubt an important finding, the discussion above indicates that the general criticism 6 See also Mair (2000); Kriesi (2007); Kriesi, Tresch and Jochum (2007); Kriesi et al. (2008, 2012); Risse (2015). 7 Hobolt and de Vries (2016) provide an excellent summary. 8 Other work in this area is by Schneider (2013) and Schneider and Slantchev (2014). 4

6 relates to EU governments being unresponsive to the specific preferences that publics have over individual policies. The next section develops a theoretical argument about when and how governments signal such political responsiveness to their citizens in European negotiations. Similar to the research on responsive governance in national democratic systems, I focus on the conditions under which governments represent the specific policy preferences of their citizens in the EU. The Politics of Responsive Governance My theory of responsive governance in the European Union is based on a standard political economy model. Voters and interest groups care about their welfare, and they have ideologically grounded policy positions. They want to elect governments who best represent their policy interests and who appear competent and responsive to them. I define responsiveness as the willingness and ability of a government to represent the interests of the politically relevant national electorate in the EU. Voters do not know the politicians level of responsiveness, but they could use information about the government s past responsiveness to assess the future responsiveness of that government; they vote rationally retrospective. It would be unrealistic to assume that all voters are fully informed about the past performance of their government in the EU. Nevertheless, they can use information shortcuts to acquire the necessary information at least in cases that are saliently discussed at the national level. Incumbent governments are opportunistic and want to get reelected. They try to appear responsive and represent the preferences of politically relevant groups at the national level, which includes individual voters and interest groups. Governments need to compete for aligned partisans and independent (or dealigned) voters simultaneously and their choices reflect their decisions to appeal to these groups. If European issues are salient to their national constituencies, then incumbents have grounds to believe that voters also care about how well they bargain in European negotiations and what type of policy outcomes they can achieve. But even in the national political arena it is extraordinarily difficult for incumbents to know which of the many domestic policy initiatives and outcomes are popular or unpopular amongst the domestic electorate, and which of these have electoral effects to begin with. To make matters worse, incumbents oftentimes do not have time to wait until they have the information necessary to decide whether they should appear responsive in those negotiations (Mayhew, 1974, 57). Uncertainty about which issues will be politicized poses a predicament to opportunistic incumbents; the wrong action or even inac- 5

7 tion could be detrimental to their electoral prospects. Consequently, as long as incumbent governments believe that their conduct in the EU could have an electoral effect they should try to signal responsiveness in European negotiations. The structure of European cooperation allows governments to signal responsiveness both at the input and output level. 9 At the output level, governments want to claim credit for EU policies that benefit their politically relevant constituents. They want to achieve negotiation outcomes that are in their voters (and interest groups ) best interest and claim that these policies are a result of their personal effort and ability during European negotiations. Opportunistic governments can also signal responsiveness at the input level by taking positions on issues that are in the interest of their domestic electorate. That is, governments want to signal that they represent their domestic supporters in EU negotiations. Position taking does not rely on the governments having to signal any personal role in outcomes, but rather on their willingness to support a particular position that favors politically relevant groups (in the EU sometimes portrayed as standing firm in the national interest ). Taking clear positions signals to the electorally relevant population that the government is fighting the good fight, and acting in their best interest. A credible signal of responsiveness implies that governments will make their positions public and that they defend these positions throughout the Council negotiations. 10 If there is at least some uncertainty about the electoral salience of a European policy issue then incumbents should have strong incentives to signal responsiveness because the absence of a signal of responsiveness could easily be interpreted as a signal of non-responsiveness. Only if voters did not care at all about European negotiations (or if incumbents did not believe that these issues are salient domestically) should governments have no incentives to signal responsiveness before national elections. The need to appear responsive should also be especially strong when a government is in trouble electorally. For example, a government facing adverse economic circumstances might have trouble persuading the electorate that the tough times are not due to its economic incompetence. Under adverse economic conditions, governments should have strong incentives to signal responsiveness in EU negotiations to counter some of the negative economic 9 The signals I discuss in the EU context are also integral features of responsiveness at the national level (Mayhew, 1974; Grimmer, Messing and Westwood, 2012; Grimmer, Westwood and Messing, 2014; Cruz and Schneider, forthcoming). 10 By defending their initial positions in Council negotiations governments depart from the European norm. Tough negotiation tactics usually receive a lot of attention in the European institutions and the national media, which makes them more likely to be picked up by the national media. 6

8 signals. But while it is safe to say that all governments want to appear responsive to their electorate before elections, not all governments are equally competent to take positions and to achieve negotiation outcomes in the EU that reflect the interest of their core voters. Their ability to pursue policies depends on their negotiation competence in an intergovernmental system where they have to negotiate with the other 27 EU member governments, and other assorted institutions. Two mechanisms can explain how EU governments can still signal responsiveness during election periods. The first mechanism is based on the assumption that each government fends for itself, and governments do not allow other governments to appear responsive before the elections. In these cases, governments can influence the negotiation strategies and outcomes using their formal or informal bargaining power, and their success beyond these characteristics purely depends on their actual level of competence. The second mechanism is based on the assumption that governments work in a relatively cooperative environment. Rather than having to use hard bargaining strategies to achieve their goals, reciprocal cooperation between EU member governments could also lead to electoral cycles in bargaining strategies and outcomes. That is, governments allow other governments to appear politically responsive even if they are not necessarily that competent. Both mechanisms can be at work in Council negotiations, potentially simultaneously. Whatever mechanism is at work, they both provide similar predictions about the effect of national elections on signals of responsiveness in Council negotiations. The central hypothesis that I would like to test in this paper is that EU governments that face elections at the national level are more likely to use Council negotiations to signal input and output responsiveness to their voters, in the hope of garnering additional public support. Responsive Governance in EU Negotiations To test the empirical implications of my argument, I use data on the bargaining positions and outcomes in EU legislative negotiations. The Decision Making in the European Union (DEU) data set provides information on the policy positions of the member states representatives in the Council on over 125 important legislative proposals that were negotiated between 1999 and The data offer me a unique opportunity to derive information on the bargaining strategies that governments use in European legislative negotiations (including position-defending strategies), as well as the ability of governments to assert their positions at the end of the negotiations (their negotiation success). 11 Thomson et al. (2006); Thomson (2011); Thomson et al. (2012). 7

9 The DEU researchers conducted 349 extensive, semi-structured face-toface expert interviews. These experts were recruited from the permanent representations of the member states, the European Commission, and the European Parliament. 12 They were mainly civil servants who were responsible for representing their country in the Council discussions and monitored the legislative negotiations closely. The collection of data on governments positions applies the spatial model of politics to specific controversies. For each policy issue (each policy proposal is divided into policy issues), experts were asked to indicate the policy positions initially favored by each government after the introduction of the proposal before the Council formulated its common position, as well as the positions that the governments represented in the final stages of the legislative negotiations. Variable Descriptions To measure the input responsiveness of EU members in the Council negotiations, I follow the existing literature and focus on EU governments willingness to move away from their initial policy positions during the legislative negotiations. 13 Policy positions that are announced at the beginning of the negotiations represent the official position of a government, and they provide a good representation of its national ideological interest (Thomson, 2011). These are the positions that EU governments share with their publics and with the media. They provide a benchmark for the positions that the government takes during the negotiations. The final position refers to the policy alternative the EU government defends on each issue just before a final decision is taken. Position Defense is a binary variable that takes the value 1 if governments consistently represent their initial position on a policy issue throughout the negotiations, and 0 if they are willing to compromise by changing their position on the policy issue before the final decision is taken. On average, governments only defend their initial positions in about 40% of policy issues that are negotiated. 14 The fact that EU governments compromise on almost two thirds of the legislative issues is consistent with the cooperative characterization of the negotiation process in the literature. Opportunistic politicians who face elections at home should be more likely to posture in the Council by declining to move from their initially 12 On average, these interviews lasted 1 hour and 40 minutes. The face-to-face interviews also served to assess the expertise of interviewees (Thomson and Stokman, 2006). 13 See, for example, Arregui, Stokman and Thomson (2004, 2006); Arregui (2008); Thomson (2011); Aksoy (2012). 14 I present a histogram in the appendix to illustrate the variation of position defending behavior across EU member states. 8

10 stated policy position and by achieving more favorable bargaining outcomes. Election Period is coded as 1 if an EU member has domestic legislative elections during the period of which the issue is negotiated, and 0 if not. EU governments should have incentives to engage in position-defending strategies throughout the entire legislative process, but the effect should be the strongest when elections fall within (or shortly after) the first Council reading, where most of the EU government s positions are solidified. My main model focuses on national elections that occur at any time during the legislative negotiations, but I also show in the appendix that the findings are robust (and substantively stronger) when I focus on elections that take place before the Council adopted its common position in the first reading. To measure the output responsiveness of EU members in Council negotiations, I use the inverse of the absolute distance between a member state s initial policy position and the final outcome on each legislative issue. 15 The resulting variable of Bargaining Success takes values between 0 and 100, whereby larger values imply greater bargaining success on a legislative issue. The average bargaining success across EU members and over the entire time frame is Are governments able to improve their bargaining success before national elections? I am particularly concerned with elections that take place after the adoption of the proposals, that is, when the bargaining success of a government is revealed. My main variable for Election Period is a dichotomous variable that takes the value of 1 if an election occurs within six months from the day of the adoption of the proposal, and 0 if not (the results are robust to taking into account elections within a year of the adoption). If governments expect to face elections shortly after the conclusion of negotiations they should be more likely to signal output responsiveness to their electorates by achieving better bargaining outcomes. The estimations include a battery of control variables following the common practice in the literature on European legislative negotiation behavior and negotiation outcomes. For Position Defense, I first control for whether the qualified majority voting rule is applied using Data from DEU (QMV). In addition, I control for the relative salience of the policy issue. The DEU data set codes the salience that a government attaches to each legislative issue. Experts were asked to estimate the level of salience for each issue on a scale from 0 to 100, with 100 indicating that an issue is of the highest importance to a stakeholder while a score of 0 indicates that the issue is of no importance whatsoever to a stakeholder. Since salience is measured as a relative concept, I follow Arregui and Thomson (2009) and measure it 15 See, for example, Bailer (2004, 2006); Arregui and Thomson (2009). My results are robust to using the weighted measure of Bargaining Success (Cross, 2012). 16 The appendix provide box plots to illustrate the variation of Bargaining Success across EU member states. 9

11 as the absolute distance between the salience that the EU member attaches to any given issue and the average salience that all EU members attach to the issue (excluding the EU government under observation). Voting Power (%) measures the formal bargaining power of an EU member in the Council using the Shapley Shubik index (SSI) (Shapley and Shubik, 1954). The SSI score analyzes the number of times an EU member is pivotal in a coalition (how often does it turn a losing coalition to a winning coalition), taking into account the votes of the government as well as the decision-making rules (majority voting vs unanimity voting). Higher values on the SSI index implies greater formal voting power. Data are from Bräuninger and König (2005). GDP (log) is measured as the annual logged gross domestic product of each EU member. Data are from Eurostat. Distance from Commission and Distance from EP measure the distance of an EU member s position to the position of the Commission and the European Parliament on each issue, respectively. Higher values on the variable indicate a greater distance between the EU member and the Commission. Finally, Position Extremity measures the distance of an EU member s position to the average position in the Council. For each EU government, I calculate the average position in the Council without the position of that government. Larger values on the variable indicate that the EU member has a more extreme position on the legislative issue than the average EU member in the Council. Data are from the DEU data set. For Bargaining Success, I use similar control variables but also include a variable on Distance from the Status Quo and the variable Position Defense. Descriptive statistics for all variables in both specifications are presented in the appendix. Empirical Results Table 1 provides a summary of the estimation results for the governments ability to signal responsiveness to their electorate before elections. Model 1 focuses on the signal of input responsiveness (Position Defense). Since the dependent variable is dichotomous, I estimate the model using the probit estimator. Model 2 focuses on the signal of output responsiveness (Bargaining Success). I estimated the model using the OLS estimator. I include country fixed effects in both specifications to control for potential unobservable confounding factors that are specific to the EU member country. Since the observations are not independent from each other, I compute standard errors that are clustered at the issue level. In an appendix, I show that the results are robust to a number of different model specifications. I estimate models that include proposal-fixed effects, proposal-clustered standard errors, additional control variables (i.e. Council type, multidimensionality of the proposal), different operational- 10

12 izations of the dependent variable, and I conduct placebo tests. I also estimate multi-level mixed-effects (probit and linear) regression models to account for the hierarchical nature of the data. Because of space limitations, these and more robustness checks are presented and discussed in the appendix. (1) (2) Position Defense Bargaining Success Election Year 0.272** 3.287** (0.111) (1.186) Relative Saliency (0.002) (0.065) QMV (0.252) Voting Power (%) ** (0.078) (1.374) GDP (log) ** (0.458) (10.133) Distance from EP ** (0.003) (0.093) Distance from Commission ** (0.002) (0.092) Position Extremity * ** (0.005) (0.113) Position Defense ** (4.572) Distance from Status Quo (0.085) Constant ** (5.785) ( ) Country FE Yes Yes Wald/F-test 90.58*** 13.37** Observations DVs: Position Defense (1) and Bargaining Success (2) Specification: Probit with Country FE (1) and OLS with Country FE (2) Issue-level clustered standard errors in parentheses * p<0.10, ** p<0.05 Table 1: National Elections and Signals of Responsive Governance. Both models fit the data well. We can reject the null hypothesis that jointly the coefficients are equal to zero for both models. Turning to the substantive findings, EU governments aim to signal both input and output responsiveness to their electorates before elections. EU governments that face an election during legislative negotiations are significantly less likely to compromise on their initial bargaining position (Model 1). The predicted probability that an EU member defends her position during negotiations increases from 0.36 when she does not have national election to 0.46 if she has a national election during the negotiations, holding all other variables 11

13 at their means. To put this differently, EU governments are less than average likely to defend their position when there is no election (the likelihood of position defense is 36%), but higher than average likely to posture when there is an election (the likelihood of position defense is 46%). Similarly, EU governments are significantly more likely to achieve successful outcomes when legislative issues were adopted in election periods. Electoral cycles in bargaining success exist if elections occur shortly after the adoption of the legislation. During election periods, governments increase their bargaining success by about 3.3%. Since 75% of sample observations fall between 50 and 100, governments de facto increase their bargaining success during election periods by almost 7%. In addition to the direct effect of national elections on bargaining success, I find strong support for an indirect influence of national elections on EU governments bargaining performance. Governments that defend their positions during the legislative negotiations (oftentimes because of national elections) are significantly more successful in legislative negotiations. 17 EU governments that defend their positions during legislative negotiations increase their bargaining success by almost 11%. Taken together, elections can exert strong indirect and direct effects on legislative policies in the EU. Greater bargaining success during electoral periods implies that policies shift for electoral purposes into directions that they would not have shifted to without one or several countries experiencing a national election. So far, the analysis indicates that EU members are more likely to signal responsiveness in the shadow of national elections. If EU governments experience national elections they are much less likely to compromise on their initial policy positions, and they are more likely to achieve successful bargaining outcomes. Theoretically, electoral cycles should be conditional on the motives and opportunities of EU governments. For example, governments should have greater incentives to induce electoral cycles if they face high unemployment rates (or other forms of economic distress) at the national level, pressuring them to demonstrate competence in other areas. Similarly, signaling responsiveness should be more likely the more willing other EU members are to cooperate with each other. For example, as the number of elections that fall within the negotiation period increases it be- 17 Note that position defense implies that governments do not change their position from their initial pre-negotiation position to the final position shortly before the final decision is taken. While they might still represent their initial position, they could be outvoted in the Council, which would imply a final outcome further away from their position. Since Bargaining Success is calculated on the basis of the final outcome position defense therefore does not determine better negotiation outcomes. The positive outcome therefore rather speaks to the importance of cooperative and consensual bargaining where governments try to accommodate each others interests in the final outcome. 12

14 comes more difficult to achieve superior bargaining outcomes because there are more countries that negotiate harder to signal responsiveness at home. Not everyone can be accommodated. Marginal Effect of Election on Position Defense Kernel Density Estimate of Unemployment Rate Marginal Effect of Election Year on Bargaining Success Mean of Unemployment (%) Kernel Density Estimate of Unemployment (%) Unemployment Rate (%) Thick dashed lines give 90% confidence interval Thin dashed line is a kernel density estimate of Unemployment Rate Unemployment (%) Thick dashed lines give 90% confidence interval. Thin dashed line is a kernel density estimate of Unemployment (%). (a) Position Defense Economic Distress (b) Bargaining Success Economic Distress Marginal Effect of Election on Position Defense Kernel Density Estimate of Number of Elections Marginal Effect of Election Year on Bargaining Success Mean of Number of Elections Kernel Density Estimate of Number of Elections Number of Elections Thick dashed lines give 90% confidence interval Thin dashed line is a kernel density estimate of Number of Elections Number of Elections Thick dashed lines give 90% confidence interval. Thin dashed line is a kernel density estimate of Number of Elections. (c) Position Defense Number of Elections (d) Bargaining Success Number of Elections Figure 1: Conditional Electoral Cycles in Political Responsiveness. Although it is beyond the scope of the paper to provide an exhaustive analysis of how motives and opportunities affect responsive governance in the EU, Figure 1 presents results of estimations that analyze the extent to which the electoral effect depends on national unemployment and the number of elections. 18 Both represent important indicators of motives and opportunities for government incentives to govern responsively. To capture these conditional effects, I re-estimate the main models and include an interaction term. 19 Since the coefficients cannot be interpreted straightforwardly, I present the marginal effect of elections on position defense graph- 18 Full estimation results are available in the appendix. 19 To generate the interactive graphs for the probit model I adapted the Stata code provided by Matt Golder. last accessed: October See also Thomas Brambor and Golder (2006); William Berry and Milton (2012). For 13

15 ically. For example, Figure 1(a) graphs the coefficient of Election Period on Position Defense (solid line; the size of effect can be read from the left y-axis) for different levels of Unemployment (x-axis). The thick dashed lines give the 90% confidence intervals, and the thin dashed line is the Kernel Density estimate of Unemployment to provide information on its sample distribution. The graphs indicate that electoral cycles in input competence (position defending behavior) do not depend on the economic distress of a government before an election or the number of elections during the negotiations. These results stand in contrast to the electoral cycle in output competence where I find that signals of bargaining success before elections are indeed more likely when unemployment is relatively high (above the EU average), and when there are not too many elections right after the conclusion of negotiations. These findings lend further support to the theoretical mechanism. The ability of EU governments to signal bargaining success depends on the negotiations of a number of actors with potentially different objectives. When defending their official positions, EU governments are less constraint. They can unilaterally decide not to compromise and cannot easily be forced to change their position (they can only be outvoted in the final step). Since governments do not have perfect knowledge about which issues might be important, it is not surprising to observe that they tend to generate unconditional electoral cycles when they are not constraint by other governments or institutional rules. The findings for the other control variables also have some interesting implications. EU members that take extreme positions relative to the average position in the Council (Position Extremity) are not only less likely to be able to defend their positions throughout the decision-making process, but they also tend to be less successful overall. The further the country s position from the European Parliament (Distance from EP), the less likely it will defend its position during the negotiations (I do not find an effect on bargaining success). For bargaining success, I also find that large countries (in respect to GDP and to population size), politically powerful countries (in terms of the Shapley Shubik voting power index), and countries that take positions further away from the Commission s position on the legislative issue are less successful. These results are in line with much of the literature on decision-making in the Council (Thomson et al., 2006; Arregui, 2008). Overall, the findings provide evidence that before national elections EU governments are more likely to defend their initial positions in order to signal input responsiveness, and they are more likely to achieve more favorable negotiation outcomes to signal output responsiveness. When cooperatthe interactive graphs for the OLS model I adapted the Stata command GRINTER provide by Fred Boehmke. 14

16 ing at the EU-level, governments are responsive to their national electorates similar to patterns of responsiveness in the national political arena. The likelihood that EU governments can claim credit for successful bargaining outcomes is thereby constraint by their need to provide such signals, and their ability to achieve these outcomes at the EU-level. How Do Voters Respond to Government Signals? Evidence from a Conjoint Experiment in Germany The theory of responsive governance is based on the assumption that the supply of political responsiveness depends on the governments expectations about corresponding demands at the domestic level. Whereas not all policies are salient at the national level, and they do not have to be in order to cause electoral cycles in political responsiveness, theoretically I would expect that voters care about the different signals of responsiveness at least when those policies are salient. But even though the literature has made much progress in analyzing whether European voters care about European integration (i.e. whether they are favorably or unfavorably disposed towards the EU), we have no knowledge whether voters would interpret position-taking, position-defending, and credit-claiming activities of their government at the EU level as signals of responsiveness. It is questionable even for highly salient European policies whether voters simply care about the negotiation outcomes (which could influence their support for the EU as an institution) or whether they also care about signals of political responsiveness (which would affect their vote choice at the national level, and in turn, their governments political responsiveness at the EU-level). Since this assumption is central to my theory of responsive governments, I now analyze this demand-side of political responsiveness: that is, how voters respond to different signals of political responsiveness when policies are salient. 20 To examine how voters respond to different signals of political responsiveness, and to assess the internal validity of the demand-side of political responsiveness, I conducted a large-scale online survey about voter responses to different signals of political responsiveness in two salient policy areas whether the EU should agree to another financial rescue package for 20 Given the data limitations, it is not possible to provide a test of whether voters rewards their governments for positive negotiation outcomes; I have provided such a test elsewhere (Schneider, 2016). In fact, I do not expect that voters always respond to signals of political responsiveness (i.e. depending on the salience of the issue), but when they respond I expect them to reward (punish) governments for being responsive (non-responsive) both at the input and the output level. In EU budget negotiations, a highly salient policy area, higher budget receipts in EU negotiations increase government approval at the domestic level at least since the 1990s. 15

17 Greece and whether the EU should allow for more immigration of refugees and asylum seekers into the EU in Germany. Both issues are currently discussed in the EU, and are not only highly politicized in Europe, but particularly in Germany, which shoulders the largest share of the burden on both financial rescues and immigration. The fact that both policies are politicized domestically allows me to test whether respondents rewarded signals of responsiveness in ways that correspond with the underlying theoretical argument. The survey was fielded in the fall of 2016; the sample includes 2450 German adults who are eligible to vote in federal elections. Although the focus of my survey experiment is on internal validity, the online sample was somewhat skewed towards younger, more educated, and male voters compared to the total voter population. To address this issue, I use entropy balancing to re-weight the data from the survey such that it matches the demographic margins from the voter population. 21 The details of the survey and the conjoint experiment are described in the appendix, but I provide a brief overview in the next section. Data and Conjoint Experiment I designed a fully randomized conjoint experiment to examine how voters assess different politicians based on the possible multidimensionality of their position-taking and position-defending strategies, and their bargaining success. 22 All respondents were instructed about the conjoint exercise and then exposed to comparisons between two politicians, each of whom varied along six different dimensions. My experimental design is modeled on previous experiments on political repositioning and voter behavior in American politics (Butler and Powell, 2014; Houweling and Tomz, 2016a,b). I proceeded in three steps. First, I asked respondents to indicate their opinions about the two issues under observation to establish a baseline on which they judge politician s behavior. 23 Respondents were presented with information on financial rescue packages to Greece, and then asked whether 21 The survey was implemented by Respondi over the internet on samples of the adult vote-eligible Germans. Although Respondi uses various techniques to achieve a sample that resembles the underlying population, online samples are never true probability samples. But whereas probability sampling has been proven more effective, it is extremely difficult (if not impossible) to achieve even with offline sampling methods, it comes at considerable costs, and its benefits diminish with historically declining response rates. Since my focus is on internal validity, I therefore opted for an online survey, and weighted the data by important demographic characteristics. Overall, the imbalances are relatively minor and the results are robust when I use the unweighted data. 22 Hainmueller, Hopkins and Yamamoto (2014) provide a formal exposition and a discussion of the method. 23 The full text of the questions are presented in the appendix. 16

18 they are for or against further financial aid to Greece. 24 Respondents were also presented with information about the current situation on the immigration of refugees and asylum seekers, and then asked whether they are in favor of accepting more or fewer refugees in the EU. 25 Responses varied from strongly in favor, somewhat in favor, neither in favor nor against, somewhat against, and strongly against. Number of Responses (%) Strongly in Favor Somewhat in Favor Neither/Nor Somewhat Against Strongly Against Number of Responses (%) Strongly in Favor Somewhat in Favor Neither/Nor Somewhat Against Strongly Against Support for Financial Rescue Package Support for Immigration (a) Greek Bailout (b) Refugee Inflow Figure 2: Support for a Greek Bailout and the Inflow of Refugees in Germany, Figure 2 provides information on the respondents attitudes toward providing another financial aid package to Greece (Figure 2(a)) and toward accepting more refugees in the European Union (Figure 2(b)). Overall, Germans in the survey are slightly more opposed to both, providing another financial bailout and accepting more refugees, but the extent of support and opposition varies across policy fields. Only 3% of Germans are strongly in favor of another financial package for Greece (6% are in favor of accepting more refugees). 18.5% of Germans are somewhat in favor of both policy initiatives. 24% are neither in favor nor against another financial package for Greece (the percentage is 21% for accepting more refugees). About 34% are somewhat against another financial bailout of Greece (28% are somewhat against accepting more refugees), and 20% are strongly against another Greek bailout (27% are strongly against accepting further refugees). 24 I use the terms bailout and financial package simultaneously. In the survey, I exclusively used the term financial rescue package (Finanzhilfe) because the term bailout tends to carry negative connotations. 25 Strictly speaking, the terms refugee, immigrant, and asylum seeker refer to different groups. In the survey, I decided to use the German word for refugee (Flüchtling) even though the current crisis centers around asylum seekers who illegally enter the borders of the EU. However, the media tends to refer to them as refugees, and much of the public debate uses the term in this matter. My discussion will therefore use the words immigrants and refugees interchangeably to refer to both refugees and asylum seekers. 17