International Treaty Ratification and Party Competition: Theory and Evidence from the EU's Constitutional Treaty

Political Science Research and Methodshttp://journals.cambridge.org/RAM

Additional services for Political Science Research andMethods:

Email alerts: Click hereSubscriptions: Click hereCommercial reprints: Click hereTerms of use : Click here

International Treaty Ratication and Party Competition:Theory and Evidence from the EU's Constitutional Treaty

Andreas Dür and Nikitas Konstantinidis

Political Science Research and Methods / Volume 1 / Issue 02 / December 2013, pp 179 - 200DOI: 10.1017/psrm.2013.17, Published online: 08 November 2013

Link to this article: http://journals.cambridge.org/abstract_S2049847013000174

How to cite this article:Andreas Dür and Nikitas Konstantinidis (2013). International Treaty Ratication and PartyCompetition: Theory and Evidence from the EU's Constitutional Treaty. Political Science Researchand Methods, 1, pp 179-200 doi:10.1017/psrm.2013.17

Request Permissions : Click here

Downloaded from http://journals.cambridge.org/RAM, IP address: 178.175.130.138 on 16 Dec 2013

Political Science Research and Methods Vol 1, No. 2, 179–200 December 2013

r The European Political Science Association, 2013 doi:10.1017/psrm.2013.17

International Treaty Ratification and Party Competition:Theory and Evidence from the EU’s Constitutional Treaty*

ANDREAS DUR AND NIKITAS KONSTANTINIDIS

What explains a party’s dual decision about whether to endorse a referendumon an international treaty and whether to support that treaty in a referendumcampaign? Relying on an original game of second-order electoral competition, this

article argues that the relative likelihood of a party endorsing a referendum is highest at thebeginning and end of the electoral cycle, and when the public supports the treaty. The study usesdata on the position of 175 parties in 24 member states vis-a-vis the EU’s Constitutional Treaty andits preferred mechanism of ratification to test these expectations against empirical evidence. Using amultinomial logistic regression model, it finds robust support for the argument.

The last two decades have seen a substantial number of facultative referendums forthe ratification of international treaties. The most prominent examples are thereferendums for the ratification of institutional reforms in the European Union

(EU). In 2005 and 2006, for example, France, Poland, the United Kingdom and severalother countries either held (or planned to hold) referendums on the EU’s ConstitutionalTreaty (also known as the Treaty Establishing a Constitution for Europe [TECE]).Interestingly, other EU member countries—among them Austria, Belgium and Sweden—decided not to submit the treaty to a popular vote. Even more variation exists whenlooking at party positions: in Spain, all political parties represented in the nationalparliament called for a referendum, while in Latvia all political parties backed ratificationin parliament. In 15 member states, political parties were split on this issue. AcrossEurope, about half of the parties that favored a referendum also supported the treaty,whereas the other half opposed it. What explains this variation across party positions withrespect to the desirability of an international treaty per se as well as of a non-requiredreferendum as an instrument of its ratification?We develop an original game-theoretic model to answer this question. A key prediction

derived from the model is that, all else equal, parties are most prone to support a referendumon an international treaty at the beginning and toward the end of the electoral cycle. At thesame time, the few parties that endorse a referendum in the middle of the electoral cycle arehighly likely to come out in favor of the international treaty. These findings come about for

* Andreas Dur is professor of international politics, Department of Political Science, University ofSalzburg, Rudolfskai 42, 5020 Salzburg, Austria ([email protected]). Nikitas Konstantinidis isTemporary University Lecturer in International Relations, Department of Politics and InternationalStudies, University of Cambridge, Alison Richard Building, 7 West Road, CB3 9DT Cambridge, UnitedKingdom ([email protected]). We are grateful to Leonardo Baccini, Christian H. C. A. Henning, ThomasSattler, Bernd Schlipphak, Gerald Schneider, Robert Thomson, Oliver Treib, Johannes Urpelainen, andparticipants at the 7th SGIR Pan-European International Relations Conference in Stockholm, the 4th AnnualConference on the Political Economy of International Organizations in Zurich, and the 1st AnnualConference of the European Political Science Association for helpful comments on an earlier version of thispaper. To view supplementary material for this article, please visit http://dx.doi.org/10.1017/psrm.2013.17

two reasons. First, voters’ incentives to treat referendums as second-order elections and usethem to signal their (dis)satisfaction with the government’s performance to date are highest inthe middle of the electoral cycle. This is what we call the ‘‘voter signaling’’ effect. Secondly,changes in their political capital as a result of ratification decisions carry greater weight forparties the closer they find themselves to the next election. We call this the ‘‘reputational’’effect of referendums.The effect of timing, however, is conditional on public opinion towards the treaty in a

country. We expect the probability of a party calling for a referendum to fall over the courseof the electoral cycle in countries with a favorable public opinion toward the treaty and togrow in countries with a skeptical public opinion. Public support for a treaty, of course,increases the likelihood that a party will support the treaty in a referendum campaign.We test our argument against a dataset that includes the positions of 175 parties on

whether to support a referendum on the EU’s Constitutional Treaty (signed in October2004) and on whether to back the treaty in a referendum campaign. Since the signing ofthis treaty fell randomly within the electoral cycle of EU member states, we can test ourargument in a quasi-experimental manner. Using multinomial logistic regression analysis,we find solid support for our theory.In developing this argument, we build on a series of recent studies that address the

question of why political parties sometimes call for optional referendums on salient issues(Schneider and Weitsman 1996; Hug 2004; Closa 2007; Hug and Schulz 2007; Tridimas2007; Finke and Konig 2009; Dur and Mateo 2011). These studies stress factors such asthe role of institutional rules (Hug 2004; Finke and Konig 2009), voters’ preferences(Tridimas 2007; Dur and Mateo 2011), parties’ expected gains from the policy (Finke andKonig 2009), the closeness of elections (Dur and Mateo 2011) and imitation acrosscountries (Closa 2007). Our main contribution to this literature is a focus on partystrategies rather than country-level outcomes. Moreover, we highlight the importance ofdomestic political factors within various institutional settings. A further novelty in ourapproach is that we simultaneously derive parties’ two-pronged equilibrium strategies onwhether to endorse a referendum and what position to take in a referendum campaign,thus obviating the need for arbitrary assumptions on party stances, which are often usedin the literature.We also contribute to the growing literature on second-order elections. For some

time, scholars have debated the extent to which voters use electoral contests ofsecondary importance to send signals to their government.1 Schmitt (2005) empi-rically establishes a curvilinear effect of the time elapsed between any EuropeanParliament election and a national parliamentary election on the degree of voteshare congruence at both the party and country levels, and presents this as evidenceof strategic voting. Our results suggest that parties take such a curvilinear effect forgranted in their decision-making process. In the same vein, Ray (2003a) uncovers apositive relationship between incumbent support and pro-EU attitudes that appearprimarily when referendums are held on the reform of European institutions or duringEuropean Parliament election years. Both our theoretical and empirical results essentiallyreplicate a similar inverted-U relationship in the case of referendums for internationaltreaty ratification.

1 For the case of referendums, see Schneider and Weitsman 1996; Svensson 2002; Garry, Marsh, andSinnott 2005 and Hobolt 2009. For the case of European elections, see Reif and Schmitt 1980 and Marshand Mikhaylov 2010.

180 DUR AND KONSTANTINIDIS

A PARTISAN MODEL OF INTERNATIONAL TREATY RATIFICATION

Basic Set-up

This article focuses on the ratification subgame, seeking to explain variation in party stanceswith respect to the use of popular referendums as ratification mechanisms. Our interest liesprimarily in cases of international treaty ratification in which the government maystrategically exercise the option of submitting the international agreement to a popular vote.In these cases, opposition parties can also position themselves strategically with respect tothe use of referendums for purely rhetorical reasons, even without having referendum-initiation powers. Incumbent and opposition parties then play a reputational game ofelectoral competition, in which the political stakes are a function of timing within theelectoral cycle. This novel theoretical approach highlights the domestic strategic contours oftreaty ratification by allowing for the probabilistic occurrence of failed referendums (as inthe case of the French and Dutch rejections of the EU’s Constitutional Treaty).Consider what happens once the negotiation stage of a far-ranging and politically

salient international treaty (such as accession to or deepening of a highly institutionalizedinternational organization) has been successfully completed. Let xc 2 X � R (X closedand compact) denote the outcome of international treaty negotiations along the singledimension of political (dis)integration and policy (de)centralization. Without loss ofgenerality, assume that the treaty negotiation outcome xc lies to the right of the status quoxc 4 xSQ� �

in terms of further deepening the existing cooperation arrangements. In thatcase, the incumbent party needs to decide whether to ratify the negotiated agreement viaparliament (with ratification assumed to be certain) or by referendum (the outcome ofwhich is uncertain). We go on to model the role of political timing and related strategicconsiderations with respect to these choices using a probabilistic game of electoralcompetition. This approach allows us to subsume the role of voters as strategic playersinto an endogenously derived probability function for the outcome of a referendum vote.We effectively restrict the zone of acceptable agreements at the intergovernmental

negotiation table to those that are weakly preferred to the status quo by boththe incumbent (Du xc;xSQ; xI

� �4 0, where xI denotes the incumbent’s ideal policy) and

the average voter (DW xc;xSQ� �

� 0, where W(�) denotes aggregate welfare) (seeAssumption 1 in the Appendix).2 Therefore, neither the office-seeking (those interestedin aggregate welfare maximization) nor the policy-seeking (those weighting particularisticinterests more heavily) factions of the incumbent will have reason to object to the treaty.In the case of supermajority parliamentary ratification rules,3 our assumption of certainratification implies that the treaty is acceptable to a pro-integration supermajority ofparliamentarians (who have publicly stated positions on integration policy and preferenceprofiles known to the incumbent). Although we recognize that the negotiated agreement isendogenously determined within a two-level game framework (Putnam 1988), we choose

2 The fact that the proposed treaty is generally welfare improving for the country as a whole iscommon knowledge. One may otherwise consider an extension of the model that allows for asymmetricinformation and signaling of the exact policy effects of the treaty on the electorate. In fact, Sattler andUrpelainen (2012) estimate a random utility model using microlevel data on two repeated referendums onEuropean integration to account for the political determinants of public support for integration in anenvironment of incomplete information about the policy effects of a treaty.

3 See Table 5 in Hug and Schulz (2007, 194–5) for a list of preference configurations and parliamentaryratification thresholds by country. In the vast majority of countries, there were clear pro-treatyparliamentary majorities and, hence, little concern about failed parliamentary ratification.

Treaty Ratification and Party Competition 181

to assume away the strategic importance of domestic institutional rules of ratification(parliamentary and referendum initiation rules) for the country’s bargaining leverage byfocusing on a partial equilibrium analysis of the ratification subgame.Opposition policy preferences along the integration dimension are presumably not

constrained by such conditions. Opposition rhetoric is assumed to be both purposiveand reflective of the underlying preferences of the party base. For the purposes of thebenchmark model laid out below, we consider the case of an opposition party withrelatively moderate views on integration policy; that is, its ideal integration policy is suchthat xO 2 xSQ;xc

� �. The Appendix also discusses the cases of extreme (pro- or anti-)

opposition views on integration and their implications for our theoretical results.4

Even though the dimensionality of such far-ranging treaties is effectively higher, this isnot cause for major theoretical concern as this parsimonious framework constitutes apartial equilibrium analysis of the ratification stage, which has a binary ‘‘accept-or-reject’’outcome with respect to a given policy package xc. Moreover, as the article’s emphasis ison the domestic political determinants of international treaty ratification, the potentialstrategic interdependence of ratification decisions across countries falls beyond the scopeof our partial equilibrium approach. Even though we choose to examine the domesticcontext of ratification in isolation, a new integration agreement may still only take effectonce it has been successfully ratified by all countries involved. While the utility losses offailed ratification in any one country are certain—since the entire agreement willcollapse—the domestic policy gains of successful ratification are diluted by the risk ofinternational agreement failure. This could explain why most pronouncements about thetiming of referendums are intentionally vague. Even so, in the empirical section we arguethat the significance of these sequencing and spatial interdependence spillover effects isempirically tenuous, since the ratification decisions were taken without prior knowledgeof the scheduled timing of referendums in other countries, and such positions wereconsidered binding in the future.

A Referendum as a Probabilistic Voting Game

Citizens care both about the policy content of the negotiated treaty and the valence ofparties.5 Valence here refers to the electorate’s changing perceptions of a party’scompetence and overall record of performance. It is independent of ideology or policyarea; instead, it is influenced by the process and outcome of ratification. Hence, partieseffectively compete along two dimensions. This set-up is akin to valence models that havebeen increasingly applied to the study of electoral competition.6 We assume that partypositions and actions in the ratification process are observed by both domestic (voters)and international (treaty cosignatories) audiences. Moreover, we model valence gains and

4 Note that the model assumes that public policy preferences related to the proposed treaty are formedindependently of the positions of political elites. Ray (2003b), however, shows that party positionsconditionally influence voter opinions about European integration. Carrubba (2001), on the other hand,emphasizes the bottom-up connection between public opinion on European integration and partypositioning. Moreover, Hobolt (2006) examines how changes in party strategies (i.e., issue framing andcampaigning) affect referendum outcomes. Finally, Gabel and Scheve (2007) and Steenbergen, Edwardsand De Vries (2007) provide empirical identification strategies for causally estimating the endogenouseffect of elite communications on public opinion with respect to support for European integration.

5 For a similar combination of issue voting and second-order concerns in an analysis of referendums,see Hobolt 2009.

6 See, for example, Schofield 2005; Ashworth and Bueno de Mesquita 2009.


losses as zero sum for both the incumbent and the opposition, insofar as they accrueto the incumbent’s current level of relative political capital. Relative political capital (d) inthis sense (that is, as a state variable of accumulated valence) connotes the overall levelof trustworthiness and popularity enjoyed by the incumbent vis-a-vis the opposition andvice versa.Democratic political elites are also motivated by office and policy. The relative

importance of these two objectives depends on the salience of the valence component tovoters (g). This salience factor depends on the time until the next election and the politicalclimate at the time of ratification, namely the level of polarization (both rhetorical andlegislative). The relative salience variable is also conditioned by the institutional andstructural features of the party system that make it either more adversarial (e.g., UnitedKingdom) or consensual (e.g., Switzerland).We follow the majority of scholars in this field in classifying referendums for the

ratification of international treaties as second-order elections that evolve into popularitycontests. Substantive debates over international policy are often shrouded by domesticelectoral motivations and strategic posturing, subject to the various audience costs andbenefits associated with different ratification outcomes. The incumbent’s perceived abilityto govern is always on the line at every popular vote.7 Partisan supporters will alwayswelcome an opportunity to reward the incumbent at the ballot, while its opponents willsnatch the chance to voice their disapproval by all democratic means possible.Probabilistic voting allows us to derive the joint effects of relative governmentpopularity (d) and relative policy salience (g) based on the probability that thereferendum will pass. It also captures the common uncertainty and converging beliefsamong political elites about the outcome of a mass vote. This way, referendum votingequilibria are implicitly ‘‘black-boxed’’ into a probability function, enabling us to focuson the strategic properties of the ratification subgame.We provide the full mathematical derivation of the probabilistic model expounded

above in the online appendix (see Proposition 1). We also show that, in the case ofindependently and uniformly distributed random variables, the probability of successfulratification by referendum takes the following form (where m . 0 captures the aggregatelevel of uncertainty over the incumbent’s relative popularity):

P g; d;DWð Þ51

21 m

DWg

1 d� �

: ð1Þ

In other words, popular governments and desirable international treaty agreements tendto be more successful in a popular referendum, while the prominence of domestic electoralfactors and political issues tend to precipitate negative referendum outcomes. Theintuition for the first two partial effects is quite straightforward. The partial effect withrespect to the relative salience parameter (g) should be viewed in light of Assumption 1 inthe Appendix (that is, DW xc; xSQ

� �� 0). Given that the treaty would be unequivocally

endorsed by a popular majority based on its intrinsic policy merits, it would be inincreasing jeopardy of not being ratified by referendum as the vote becomes furtherovershadowed by domestic electoral issues. For the rest of the analysis, we employ thesimple specification of the probability function provided in Equation 1. In the online

7 However, party positions on foreign affairs and international co-operation do affect nationalelections. On the national electoral effects of European integration politics and referendums, see de Vries(2007, 2009) and Tillman (2004).


appendix we derive more general conditions that are necessary in order for ourcomparative statics predictions to hold; namely, that the average welfare differential oftreaty implementation for swing voters needs to be non-negative (Assumption 2).

An Incumbent-Opposition Game of Treaty Ratification

We now consider the strategic interaction between the incumbent (I) and the opposition(O) with respect to the process of treaty ratification. Our game-theoretic approach clearlycaptures the strategic interplay between the incumbent’s constitutional prerogatives andthe opposition’s rhetorical actions. We start with a normal-form specification of the game,in which the incumbent needs to decide whether to ‘‘call’’ (C) a binding non-requiredreferendum for treaty ratification8 or ‘‘not to call’’ (NC). If the government calls areferendum, it essentially opts for an uncertain lottery outcome, whereby the treaty will beratified by a popular majority according to the probability parametrically specified in theabove probabilistic voting subgame. A negative referendum outcome would imply thatthe whole ratification process will stall, and that the agreement has to be renegotiated atthe intergovernmental level. Hence, the status quo level of integration and cooperation(xSQ) constitutes the reversion point. If the incumbent does not call a referendum, thetreaty is ratified by a parliamentary majority commanded by the government. While levelsof party cohesion and parliamentary voting thresholds vary across political systems, ourassumption of certain parliamentary ratification is plausible within the context ofparliamentary European democracies with high levels of party cohesion.While the incumbent naturally always comes out in favor of the treaty agreement it

negotiated multilaterally, the opposition’s strategic posture is more nuanced, as it needs todecide where it stands both with respect to the treaty itself and the mechanism ofratification. Depending on its own policy preferences xO

� �, among other things, it may

choose to ‘‘endorse’’ a referendum and ‘‘come out against’’ the treaty (EN), to ‘‘endorse’’and ‘‘come out in favor’’ (EY) or ‘‘not to endorse’’ a referendum in the first place (NE).Failure to endorse a referendum need not be qualified by a specific stance with respect tothe treaty itself, as it implies tacit consent to the ratification of the treaty and its continuedacceptance in the future as part of the country’s international commitments. In light ofour assumption about party cohesion and simple majority rules9—and the assertion thatvoters only care about policy outcomes insofar as they are excluded from the ratificationprocess—the position and rhetorical intensity of the main opposition party in parliamenthas no effect on the payoffs of the game or on the hypothetical outcome of thereferendum vote. Since parliamentary debate is presumed to be insulated from externalaudiences, parties are assumed to vote purely on policy grounds (i.e., they vote ‘‘yes’’ ifand only if Du xc;xSQ; xi

� �� 0).

8 In many countries, referendums on international treaties are not formally binding. However thedistinction between consultative and binding referendums seems irrelevant, as it is hard to imagine agovernment acting against the will of its people as expressed in a referendum. That is, the audience costsof overruling a majority of voters are considered prohibitive. On this point, see Setala (1999), Tridimas(2007, 677), Hobolt (2009, 10) and Trechsel (2010, 1062).

9 However, if supermajority rules are in place for parliamentary ratification (depending, for example,on the assessed compatibility of the treaty with a country’s constitutional order), then the opposition’saction space has to be qualified even further in order to account for its position vis-a-vis the treaty in theparliamentary ratification process. In this case, the valence dimension of the vote would presumably bemore salient, especially if the government’s survival were at stake. Yet extending the model to account fora finer menu of valence lotteries does not alter our core intuition and results.


Let us next lay out the payoff structure along the valence dimension. We essentiallyposit that the timing and outcome of a referendum will have a similar effect on the relativepolitical capital of the major political parties as a mid-election poll. In the event ofsuccessful ratification by referendum, the incumbent (I) is rewarded with a relative valencegain b 2 0; 1ð Þ for reflecting the majority will through a mechanism of direct democracy.The same applies to the opposition (O) if it supports the treaty. On the other hand,c (normalized to unity) denotes the relative loss in popularity for going against thepopular will of the ex post majority. This loss will be borne by either of the mainstreamparties as long as they position themselves ex ante on the opposite side of the referendumoutcome.10 Finally, d 2 0; 1ð Þ captures the valence net reward (punishment) of giving (notgiving) voters a direct say in the treaty ratification process. This parameter may vary withrespect to a country’s constitutional tradition in direct democracy (measured, forexample, by the total number of past referendums on national and international issues).Note that while parameters b and c are contingent on the outcome of the popular vote,parameter d is not. Moreover, the asymmetry between reputational gains and lossesbo c5 1ð Þ is justified for two reasons: (1) if a failure to ratify will cause negativeintegration externalities on the other treaty cosignatories, the opprobrium of internationalaudiences (directed primarily against the incumbent government)11 may affect the party’sdecision-making calculus in order to avoid negative reputation costs and (2) standardloss-aversion arguments apply (losing hurts more than winning helps).The probabilistic nature of referendums gives rise to risky alternatives (lotteries) that

depend on the outcome of the vote. The lottery associated with the policy component ofthe utility function is P5 xc;P; xSQ; 1�P

� �; the treaty is implemented (subject to

successful ratification in all countries) at point xc with the probability P of a ‘‘yes’’ vote(as specified in Equation 1), and the status quo level of integration prevails with theprobability 1–P of a ‘‘no’’ vote. Lotteries over the orthogonal valence payoffs will dependon the opposition’s pure strategy response to the incumbent’s choice to initiate areferendum (C): (1) if the opposition chooses to endorse a referendum publicly calling fora ‘‘no’’ vote (EN), then the incumbent is faced with the lottery over valence payoffsLIEN 5 11 b;P; � 11 bð Þ; 1�Pð Þ, with an expected value of EV LI

EN

� �5 2P� 1ð Þ 11 bð Þ;

(2) if the opposition chooses to endorse a referendum publicly calling for a ‘‘yes’’ vote(EY), then the incumbent is faced with the lottery over valence payoffs LI

EY 5 0;P; 0; 1�Pð Þ,where EV LI

EY

� �5 0 and (3) if the opposition chooses not to endorse a referendum

publicly (NE), then the incumbent is faced with the lottery over valence payoffsLINE 5 b;P;�1; 1�Pð Þ, with an expected value of EV LI

NE

� �5Pð11 bÞ � 1. Since

competition along the orthogonal dimension of political capital accumulation is modeledas zero sum, the corresponding valence lotteries faced by the opposition are the same but withthe payoff signs reversed: LO

aO 5�LIaO ; 8a

O 2 EN;EY ;NEf g. Note that in the case of(C, EY), the electoral fortunes of the two mainstream parties are tied together, as theirrelative gains and losses cancel each other out.

10 Schneider and Weitsman (1996, 591) assume a similar payoff structure of reward and punishmentwith respect to referendum outcomes.

11 A characteristic example of this situation is the run of events that led to the resignation of thePapandreou government in Greece in November 2011 following the prime minister’s unilateral decision tocall for a national referendum on the EU bailout and austerity package. This caused an ireful response onthe part of the country’s EU partners in the G20 summit meeting in Cannes, as it was perceived as areckless political gamble that jeopardized the future of the Euro.


The normal form of the ratification game shown in Table 1 encapsulates all the abovefeatures of the model while remaining agnostic about the sequence of moves by the majorpolitical actors. This is generally the case in games of political rhetoric. Since subgameperfection is a refinement of Nash equlibria, the simultaneous-form version of the gameprevents us from eliminating any plausible equilibria as a result of restrictive (and possiblyerroneous) assumptions about the order of play. Quasi-linear utility with respect tointegration policy and valence implies that parties will be risk-neutral with respect tochanges in their relative political capital (hence the expected value calculation of thevalence gambles); on the other hand, risk attitudes toward changes in policy remainambiguous. Note that for very high values of salience g, the game essentially becomes oneof pure conflict as the policy component becomes less relevant.An examination of the Nash equilibria of the game (see Correspondence 4 in the

Appendix) clarifies what we mean by the strategic interplay between incumbentconstitutional prerogatives and opposition rhetoric. Essentially, the policy componentof the utility function is determined by the incumbent’s choice of ratification instrument,while the political stakes of the referendum are set by the opposition’s strategic posture.If O opts for the polarizing strategy of endorsing a referendum in opposition to the treaty(EN), then it does so with the intention of raising the stakes (or else widening the spreadof lottery outcomes). On the other hand, the strategy of supporting treaty ratification byreferendum (EY) neutralizes the stakes, since EV LI

EY

� �5 0. The lottery LI

NE resultingfrom no endorsement (NE) lies in the middle in terms of its spread of potential payoffs.If, on the other hand, the incumbent opts for (certain) parliamentary ratification, then it isalways best for the opposition to endorse a referendum and thus reap the democraticlegitimacy valence payoff d. To summarize, in the benchmark specification of our partisantreaty ratification model, the opposition will be in control of the political stakes of thezero-sum valence dimension of electoral competition.However in this version of the game, in which the opposition is assumed to have

moderate policy preferences and no referendum-initiation prerogatives per se, its policystance along the integration dimension (xO) has no strategic relevance. In other words, theopposition’s strategic posturing in the process of international treaty ratification does notvary due to its intrinsic policy preferences. Even if O were negatively predisposed towardthe new integration agreement Du xc; xSQ; xO

� �o 0

� �, it may well choose to support its

ratification by referendum (EY) as long as the average voter is sufficiently pro-integration

TABLE 1 Ratification Game with Simple Majority Referendum-Initiation Provisions

‘‘O’’‘‘I’’ EN EY NE

C Eu P;xI� �

1 gEV LIEN

� �; Eu P;xI

� �1 gEV LI

EY

� �; Eu P; xI

� �1 g EV LI

NE

� �1 d

� �;

Eu P;xO� �

� gEV LIEN

� �Eu P;xO� �

� gEV LIEY

� �Eu P; xO� �

� g EV LINE

� �1 d

� �NC u xc;xI

� �� gd; u xc;xI

� �� gd; u xc;xI

� �;

u xc;xO� �

1 gd u xc;xO� �

1 gd u xc;xO� �

Note: the incumbent (‘‘I’’) has the option of calling (C) or not calling for a referendum (NC), while theopposition (‘‘O’’) has the option of endorsing the referendum while saying ‘no’ to the treaty (EN),endorsing the referendum while saying ‘yes’ to the treaty (EY) or not endorsing the referendum at all(NE). P denotes policy-related lotteries, L denotes valence-related lotteries and g captures the relativesalience of these two orthogonal dimensions of electoral competition. Finally, xI and xO correspond tothe ideal policies of the incumbent and the opposition, respectively.


and the overall political climate is conducive to popular ratification. This apparentparadox is a direct implication of the constitutional allocation of referendum-initiationprerogatives, which in the benchmark model rest wholly with the majority party.12

Comparative Statics and Electoral Timing

Our key contribution to the study of international treaty ratification is that weaccount for the fact that the timing of ratification falls randomly within the domesticelectoral cycle of each signatory country. We treat the timing of the ratification processas the end of the negotiation process. This allows for a quasi-experimental empiricaldesign that explains a substantial part of the cross-country variation in party-levelratification strategies by controlling for domestic explanatory variables in isolation fromthe international context.13 We focus mostly on relative political capital and relativepolitical salience and treat them both as state variables (d(t) and g(t), respectively) thatfluctuate during the term subject to stochastic shocks around that trend. We surmisethat the incumbent’s relative political capital is subject to a negative decaying trendover the length of its term. On the other hand, the expected time trend of the relativesalience of valence is less straightforward and essentially non-monotonic when viewedfrom the joint perspective of voters and politicians. This exogenous variation with respectto the electoral cycle proves to be very useful for the empirical identification of ourresearch question.We proceed to analyze the model’s comparative static predictions with respect to

electoral timing by way of trend and off-trend changes in the underlying time-varyingparameters d(t) and g(t). Secondarily, we also discuss the exogenous partial effects ofother time-invariant domestic-level parameters, namely DW xc;xSQ

� �and Du xc; xSQ; xj

� �.

We apply the properties of monotone comparative statics (Ashworth and Bueno deMesquita 2006) and extract a number of interesting theoretical predictions from ourmodel. More specifically, we seek to predict the direct partial effect of each time-varyingparameter on the relative likelihood between two given actions for a player by derivingtheir effects on that player’s utility differential compared to all other actions by theopponent. If changes in the model’s underlying parameters cause the utility differentialsbetween any two actions to move in the same direction regardless of the opponent’sresponse, then one can unambiguously predict a monotonic partial effect on the relativeprobability that a player will choose one action over another. This analysis is inaccordance with the multinomial logistic regression model used in the empirical section ofthe article. The formal derivation of the comparative statics properties of the model ispresented in the online appendix.In regards to our first parameter of interest, government popularity d(t), we know from

above (Equation 1) that it is positively related to the probability of a ‘‘yes’’ vote (‘‘voter

12 See the Appendix subsection on alternative constitutional provisions for referendum initiation fordifferent specifications of the ratification game.

13 Of course, within reasonable time constraints the incumbent may choose to (a) hold a referendum atthe same time as a parliamentary election, (b) leave the issue of ratification to a government with arenewed political mandate, (c) schedule the timing of the referendum ahead of its cosignatories or (d)procrastinate. Clearly, capturing a certain game of attrition with respect to the actual timing ofreferendums falls beyond the scope of this article, as we focus on the domestic determinants of parties’decisions about whether to hold a referendum in the first place. In contrast to extant work on endogenouselection timing, we treat the timing of the ratification process as exogenous.


signaling’’ effect). Taking partial derivatives of the utility differentials (see Equation 5 inthe Appendix) for both the incumbent and the opposition leads us to predict that anexogenous off-trend increase in the incumbent’s stock of relative political capital isassociated with: (1) a positive net effect on the relative odds of endorsing a referendumand supporting the treaty vs. not endorsing a referendum for opposition parties (or calling vs.not calling a referendum for incumbents), (2) a negative net effect on the relative oddsof endorsing a referendum and opposing the treaty vs. not endorsing a referendum and(3) a positive net effect on the relative odds of endorsing a referendum and supporting thetreaty vs. endorsing a referendum and opposing the treaty. All in all, mainstream oppositionparties will not want to position themselves against popular incumbents, so they will eitherchoose to wage their battle in the parliamentary arena or—unless locked in an anti-integration stance—openly endorse both the treaty and its ratification by popular vote.14

HYPOTHESIS 1: An exogenous increase in the incumbent’s relative political capital has(1) a positive net effect on the relative odds of EY vs. NE, (2) a negativenet effect on the relative odds of EN vs. NE and (3) a positive net effect onthe relative odds of EY vs. EN.

The comparative statics behavior of the model with respect to the relative salienceparameter g(t) is more complicated simply due to the fact that g enters the utilitydifferentials both as a multiplicative factor and through the referendum-successprobability function P(�). An increase in g has a twofold effect on parties’ strategiccalculus: (1) it leads to a decrease in the probability of successful ratification byreferendum (‘‘voter signaling’’ effect) and (2) it raises the relative weight of the domesticpolitical stakes of the ratification gamble (‘‘reputational’’ effect). The ‘‘voter signaling’’effect captures the degree to which voters seek to signal their (dis)satisfaction with thegovernment’s performance to date ex ante (before the outcome of the vote is known),thereby inducing favorable changes in government policy. From the point of view ofparties, the ‘‘reputational’’ effect reflects the expected present discounted value (in termsof future electability) of changes in their stock of relative political capital ex post. In otherwords, the closer the next elections are, the stronger is the effect of off-trend fluctuationsin parties’ valence on their upcoming electoral performance. The combination of thesetwo opposing effects on the behavior of the conditional utility differentials implies astrong non-monotonic relationship between g and the relative odds of any two actions.When considering the joint effects of government popularity (decreasing trend) and

relative salience (non-monotonic trend) along the electoral cycle trend,15 we expect thatthe ‘‘voter signaling’’ and ‘‘reputational’’ effects intensify (or weaken respectively) atdifferent rates throughout the electoral cycle, thus giving rise to strong non-

14 Extending this argument to multiparty systems entails qualifying the concept of relative governmentpopularity by the number of effective parties. The political stakes of the reputational gamble inherent inthe ratification game depend very much on the political setting—that is, whether it is an adversarial two-party system or a more consensual multiparty system. Undoubtedly, the valence dimension of thispolitical contest will be more pronounced in the former setting, where the identity of the two main partyrivals is more clear.

15 While we are able to provide a direct, instantaneous measure of government popularity, we can onlyproxy for the trend effects of relative valence salience indirectly through our timing variable. Ourhypothesis about the role of electoral timing then captures the joint trend effects of both of thesevariables.


monotonicities. More specifically, the ‘‘voter signaling’’ effect will tend to be strongertoward the earlier part of the electoral cycle (as long as sufficient time remains forcorrective action on the part of a government with high expected levels of popularity),while the ‘‘reputational’’ effect will tend to predominate toward the end of the cycle, aspoliticians are bound to place more weight on the electoral impact of fluctuations in theirstock of political capital the closer they get to an electoral campaign.

HYPOTHESIS 2: The political timing of ratification has a convex non-monotonic trendeffect on the relative odds of EN vs. NE and EY vs. NE. More specifically,the net likelihood that any party does ‘‘not endorse’’ a referendumincreases in the earlier part of the government’s term, peaks in the middleand starts decreasing in the latter part of the cycle.

Another parameter of interest is the aggregate welfare differential of treaty ratificationvis-a-vis the status quo DW xc;xSQ

� �� . All else equal, we consider exogenous changes in

the aggregate policy desirability of the new integration agreement, holding its policycontent fixed at its negotiated level. We find that this variable affects utility differentials intwo ways. First, higher welfare gains from treaty ratification imply a higher probability ofa ‘‘yes’’ vote (Equation 1); hence, incumbent parties will be more likely to call for areferendum and opposition parties will be more likely to endorse one. Secondly, the morecontroversial and politicized the proposed treaty, the higher the public demand for areferendum. Bypassing popular consultation through parliamentary means on highlycontroversial international issues is quite costly for politicians across the board. When theelectorate does (not) expect to be directly consulted on grave issues of foreign policyorientation, then we expect d to be relatively high (low). Therefore we assume that themagnitude of the democratic legitimacy variable d is a decreasing function of the averagepopularity of the treaty d 0 DWð Þo 0ð Þ.Taking partial derivatives of the utility differentials (Equation 5 in the Appendix) for

both the incumbent and the opposition leads us to the theoretical prediction that the neteffect of DW on the relative odds of endorsing a referendum and supporting the treaty vs.not endorsing a referendum for opposition parties (or calling vs. not calling a referendumfor incumbents respectively) is ambiguous. Yet the net effect on the relative odds ofendorsing a referendum and opposing the treaty vs. not endorsing a referendum isnegative for opposition parties (trivially so for incumbents). Finally, the net effect ofpublic support for the treaty on the relative odds of endorsing a referendum andsupporting the treaty vs. endorsing a referendum and opposing the treaty is positive foropposition parties (again trivially so for incumbents).

HYPOTHESIS 3: Parties are more likely to endorse a referendum the more skeptical publicopinion is of the treaty that has to be ratified.

The magnitude of the effects of d(t) and DW is secondarily contingent on the time-varyingrelative salience parameter g(t). Hence, electoral timing can have second-order effects onparties’ ratification choices, which we account for empirically using a model withinteraction timing effects.In terms of parties’ exogenous ideological preferences along the integration dimension,

the ratification game in Table 1 would lead us to expect that ceteris paribus, the morevaluable the proposed treaty is along the policy dimension relative to the status quo(higher Du xc; xSQ; xj

� �), the more hesitant parties are to risk failure of ratification by


submitting it to a popular vote. This, in turn, implies lower relative odds of endorsing areferendum and supporting the treaty vs. not endorsing a referendum, lower relative oddsof endorsing a referendum and opposing the treaty vs. not endorsing a referendum, andhigher relative odds of endorsing a referendum and supporting the treaty vs. endorsing areferendum and opposing the treaty. Note that in cases where referendum-initiationprerogatives only lie with the party holding the majority of seats in parliament opposition,policy preferences are expected to have no effect on equilibrium outcomes; this is not thecase under alternative (minority or supermajority) constitutional provisions.16

Finally, in countries where instruments of direct democracy are more highly valuedper se (that is, in countries with higher values of d), we should expect both incumbent andopposition parties to be more eager to endorse ratification by referendum, regardless oftheir position on the issue in hand.Table 2 presents a list of the proposed parameters of our model along with their

predicted effects on the relative odds between any two pairs of actions with respect to thechoice of ratification instrument and endorsement of the proposed treaty.

EMPIRICAL ANALYSIS

We test our argument with respect to political parties’ decisions about whether to endorsea referendum and whether to support the EU’s Constitutional Treaty. This treaty offers aunique opportunity to test our argument in a large-N study, as political parties in all butone EU member country had to decide whether or not to support a referendum.17 While asubstantial number of referendums on other international treaties has been held,18 thecase of the Constitutional Treaty includes cases in which a call for a referendum can beobserved as well as all cases in which such a call could be observed. Although this alsoapplies to the Treaty of Lisbon (signed in 2007), collusion among European governments

TABLE 2 List of Model Parameters, Variables and Predicted Effects

Predicted Effect

Parameter Variable EN vs. NE EY vs. NE EY vs. EN

d(t) Relative political capital 2 1 1g(t) Relative valence salience Non-monotonicDW xc; xSQ

� �Public support 2 1/2 1

Du xc; xSQ;xj� �

Party benefit 2 2 1

d Legitimacy benefits 1 1 0

16 See the Appendix for further details.17 We exclude Ireland from the analysis, as referendums on EU treaty changes have been considered

mandatory in Ireland since a 1987 Supreme Court ruling.18 No fewer than 48 referendums have been held on European integration since the 1960s. Going

beyond the EU, in 2010 Slovenia convened a referendum on a border deal with Croatia and Iceland onloan agreements with the Netherlands and the United Kingdom. Several countries also convokedreferendums on membership in the North Atlantic Treaty Organization (NATO), including Spain (1986),Slovenia (2003) and Georgia (2008). A further group of countries, including Croatia, Lithuania, Serbiaand Ukraine, have witnessed debates about referendums on NATO membership. Outside of Europe,Costa Rica held a popular vote on the ratification of the Central America Free Trade Agreement in 2007and Taiwan on membership in international organizations in 2008.


to avoid referendums had a strong exogenous impact on party positioning in that case.Our dataset thus includes information on 175 parties that were represented in the nationalparliaments of 24 EU member countries in 2003 and/or 2004.

The Data

Our dependent variable (Party Position) is a nominal variable that combines the positionsof a party with respect to the Constitutional Treaty and the desirability of a referendum.It is coded 1 for parties that did not endorse a referendum (NE), 2 for parties thatendorsed a referendum and opposed the treaty (EN) and 3 for parties that endorsed areferendum and supported the treaty (EY). The data for this variable come from Dur andMateo (2011), who use a variety of sources, including party websites and direct contactswith party members, to gather the necessary evidence. According to this dataset, of the175 parties included in the analysis, 75 (42.9 percent) opposed a referendum, 46 (26.3percent) supported a referendum but opposed the treaty, and 54 (30.9 percent) supportedboth a referendum and the treaty.19

Hypothesis 1 emphasizes the incumbent’s stock of relative political capital (d) (PoliticalCapital). We use two proxies for this variable. First, we rely on Eurobarometer data(2004b) on whether respondents ‘‘tend to trust [the national government] or not to trustit’’ from a survey carried out in February and March 2004. We assume that responses tothe question on trust in government are highly influenced by respondents’ evaluation ofthe current government.20 Secondly, as a cross-check we also use data from the June 2004European Election Study (European Election Study 2009). This survey included thequestion: ‘‘Do you approve or disapprove the government’s record to date?’’ Our variableis the share of valid responses that indicates approval of the government; it is highlypositively correlated with the Eurobarometer data (r5 0.71).Hypothesis 2 stresses a referendum’s timing within the electoral cycle (Timing). We

measure this variable in days remaining in the electoral term as of 1 January 2004.21 While insome countries governments can decide (or are forced) to call elections early, in general thelength of the electoral term as specified in the constitution is a good proxy for the actuallength of term. The 1 January 2004 cutoff date is based on the reasoning that it was aroundthat time that most parties made a decision on whether to support a referendum (the partiesmoving first took a decision in mid-2003, those moving last in mid-2004).22 Since none of thecountries in the sample held an election in the fall of 2003 and only three did so in the firsthalf of 2004, our results for this variable are not sensitive to the precise cutoff date chosen.We take account of our theoretical expectation of a non-monotonic influence of g on parties’decision making by also including the square of Timing in our empirical model.Hypothesis 3 draws attention to the role of public opinion (Public Support). Data for

this variable come from a Eurobarometer poll from January 2004 (Eurobarometer 2004a)

19 We excluded the Bundnis Zukunft Osterreich from the Dur and Mateo dataset, as this party wasformally launched only in 2005.

20 The results remain essentially the same when using change in trust between Spring 2003 and Spring2004.

21 We divide this number by 100 to be able to present coefficients within two decimal places.22 See Hug and Schulz (2007, 187); Finke and Konig (2009, 352). Unfortunately, it is prohibitively

difficult to have systematic data concerning the exact date on which individual parties decided to endorseor oppose a referendum. This makes a quantitative test of the argument stipulating strategicinterdependence across parties in different countries impossible.


that asked respondents to state whether they agreed or disagreed with the statement that‘‘the European Union must adopt a Constitution.’’23 While this question did not directlyask respondents about their evaluation of the Constitutional Treaty (which at that timewas not yet finalized), the responses give an indication of the project’s support amongvoters when parties decided whether to call a referendum.24 The model assumption thatDW xc; xSQ

� �� 0 is corroborated by the fact that the minimum value of public support

for the treaty was reported to be 51 percent in the United Kingdom. Below, we also testwhether the effect of Public Support is conditional on Timing.Using public opinion to explain party positions begs the question of the direction of

causality: does public opinion influence party positions, or do party positions influencepublic opinion? In the absence of an instrumental variable that predicts public opinionbut has no independent effect on party positions, we cannot fully resolve this issue.Nevertheless, the fact that the Constitutional Treaty was not a particularly salient topicfor the general public in January 2004, when the data were collected, eases theendogeneity concern. In fact, in November 2003 only 29 percent of respondents to asurvey knew that the Convention on the Future of Europe had prepared a draftConstitutional Treaty (Eurobarometer 2003a, 12).We also include several control variables in the empirical analysis below. First, we control

for a party’s general stance on European integration (Party Benefit) relying on hand-codedparty programs for the 2004 European Parliament elections (Braun et al. 2007; Veen 2011).Secondly, we operationalize the democratic legitimacy benefits of a referendum (d) using aEurobarometer poll from Spring 2003 (Eurobarometer 2003b) that asked respondentswhether they considered it essential, useful but not essential, or useless ‘‘that all citizens ofthe European Union could give their opinion, by referendum, on the draft Constitution’’(Legitimacy). Thirdly, the variable Minority takes the value 1 for parties in the CzechRepublic, Denmark and Slovenia—that is, countries in which a parliamentary minoritycould force the government to hold a referendum on the Constitutional Treaty either byrefusing to accept parliamentary ratification (in countries with a qualified majorityrequirement for parliamentary ratification) or by using constitutional provisions thatallow a minority of parliament to call a referendum. Fourthly, New Member is a dummyvariable for countries that acceded to the EU in May 2004. Fifthly, we operationalize thecompetitiveness of the political system (Competitiveness) using the number of effectiveparties (that is, the number of parties weighted by their fraction of votes) in the last electionsheld before the start of the intergovernmental conference (Gallagher and Mitchell 2008).Finally, we include a dummy variable for left-wing and liberal parties, which tend to be moresupportive of direct democracy—and, therefore, of referendums—than right-wing parties(Ideology). The Appendix offers more information on the control variables and summarystatistics for all variables included in the models.

Testing the Argument

As our dependent variable is nominal with three categories (not endorse (NE); endorse,treaty no (EN); and endorse, treaty yes (EY)), we use multinomial logit regression to

23 The variable included in the model is the proportion of supporters among the respondents thatindicated an opinion.

24 In fact, at the country level responses to this poll are highly correlated with responses to thefollowing question, which was asked in November 2004: ‘‘Based on what you know, would you say thatyou are in favor of or opposed to the draft European Constitution?’’ (r5 0.75) (Eurobarometer 2005).


estimate our models.25 The coefficients estimated using multinomial logit models capturethe increase or decrease in the log odds of being in a specific outcome category (in ourcase, for example, the probability of EN relative to NE), given a one-unit change in thepredictor (Long and Freese 2006). The coefficient for the EY vs. EN comparison is thedifference between the estimated coefficients for the EY vs. NE and EN vs. NEcomparisons, and therefore is not listed separately in the tables below.The multinomial logit model relies on the independence of irrelevant alternatives

assumption. For three reasons, we argue that this assumption does not pose a problem forour analysis. First, formal tests (Hausman and Small-Hsiao) fail to reject the assumptionfor our models. Secondly, the recent literature shows that the multinomial logit modelproduces consistent results even if the assumption is severely violated (Dow and Endersby2004). Thirdly, the results from a multinomial probit model that relaxes the independenceof irrelevant alternatives assumption are substantially the same as those reported below(see Models A1-A3 in the Appendix).Strategic interdependence, as modeled above, makes us expect little intracountry

variation in referendum endorsement. The data confirm this expectation: of 113opposition parties in our dataset, only 13 adopt a position on whether to endorse areferendum that differs from the one taken by government parties. By contrast, we expectsignificant intracountry variation in stances on the treaty. To take this into account, weinclude a cluster term in our models and further show in our robustness checks that thekey results do not change when estimating a multilevel model with random intercepts. Inmost of the models we do not distinguish between incumbent and opposition parties asincumbency status has no independent effect in our strategic game; in fact, when we dropall government parties in a robustness check, this does not change the results.The results are summarized in Table 3. Model 1 provides support for most of the

expectations derived from our formal argument.26 As expected in Hypothesis 1, the morepolitical capital a party has, the less likely it is to support a referendum but oppose thetreaty. The negative and weakly statistically significant coefficient of Political Capital forthe EY vs. NE comparison, however, runs counter to our argument. Interestingly, themore political capital a government has, the less likely all parties are to endorse the treaty.Closeness to the next election reduces the probability that a party will endorse a popularvote but oppose the treaty in a referendum campaign. Moreover, Timing2 is positive andhighly statistically significant in the EN vs. NE comparison. This finding supportsHypothesis 2 and our broader argument that parties are concerned about second-orderelection effects, which should be highest in the middle of the electoral cycle. Finally, asexpected in Hypothesis 3, the higher public support is, the less likely a party is to call for areferendum.Several of the control variables have the expected effects. The larger Party Benefit is, the

less likely the party is to support a referendum and reject the treaty. Our findings for theLegitimacy variable also make sense: the greater the demand for a referendum, the morelikely it is that parties come out in favor of both a referendum and the treaty. Parties innew member countries are less likely to endorse a referendum and support the treaty.Finally, left and liberal parties are more likely than other parties to endorse a referendum

25 Wald and likelihood-ratio tests show that no pair of alternatives in the dependent variable can becombined. Our original inclination to use a multinomial model is thus confirmed by the data.

26 Note that we lose six observations for Party benefit owing to missing data.


and support the treaty. The coefficients for Minority and Competitiveness are notstatistically significant in any of the three comparisons.A possible objection to our findings is that parties’ decisions about whether to endorse a

referendum are interdependent across countries (Closa 2007). Alas, our data do not allowus to tackle this objection head on, as we do not know exactly when each party decidedwhether to endorse a referendum (see also Footnote 22). Beyond the case of the FrenchUnion pour un Mouvement Populaire, however, we have little reason to suspect a spatialdiffusion effect influencing parties’ decision making.

Substantive Effects

The overall explanatory power of the model is very good. Model 1 correctly predicts71 percent of observations, which compares favorably with a chance result (based on themodal category in the overall dataset) of 42.9 percent. The high percentage of correctlypredicted cases is particularly noteworthy, as predicting one of three categories is more

TABLE 3 Explaining Party Positions on the Ratification of the Constitutional Treaty

Model 1 Model 2 Model 3

EN/NE EY/NE EN/NE EY/NE EN/NE EY/NE

Predictors:Political Capital 20.06*** 20.04* 20.03 20.01 20.06** 20.04*

(0.02) (0.02) (0.03) (0.02) (0.03) (0.02)Timing 21.95*** 20.41 20.01*** 20.01*** 20.00 0.00

(0.50) (0.31) (0.00) (0.00) (0.00) (0.00)Timing2 0.11*** 0.02 0.00*** 0.00*** 0.00*** 0.00

(0.03) (0.02) (0.00) (0.00) (0.00) (0.00)Public Support 20.20*** 20.07 20.33*** 20.21*** 20.19*** 20.07

(0.07) (0.07) (0.07) (0.06) (0.06) (0.07)Political Support3Timing 0.00*** 0.00***

(0.00) (0.00)Political Capital3Timing 20.00 20.00

(0.00) (0.00)

Controls:Party Benefit 20.11*** 0.01 20.11*** 0.01 20.11*** 0.01

(0.03) (0.01) (0.03) (0.01) (0.03) (0.01)Legitimacy 20.02 0.07** 20.09** 20.02 20.02 0.06*

(0.03) (0.03) (0.04) (0.05) (0.03) (0.03)Minority 1.04 1.67 1.76 1.88 1.25 1.76

(1.41) (1.36) (1.10) (1.21) (1.18) (1.10)New Member 21.61** 21.35** 21.78*** 21.69*** 21.18 20.80

(0.68) (0.59) (0.63) (0.60) (0.83) (0.72)Competitiveness 0.26 20.04 20.28 20.69** 0.12 20.11

(0.19) (0.21) (0.33) (0.32) (0.19) (0.21)Ideology 0.32 1.03*** 0.25 0.90* 0.16 0.96**

(0.58) (0.39) (0.65) (0.49) (0.58) (0.41)

Constant 26.45*** 3.10 8.93* 5.30 15.84*** 0.68(5.66) (4.91) (4.71) (5.16) (4.39) (4.67)

N (clusters) 169 (24) 169 (24) 169 (24) 169 (24) 169 (24) 169 (24)Pseudo R2 0.38 0.38 0.40 0.40 0.46 0.46BIC 338.39 338.39 315.22 315.22 337.58 337.58

Note: estimated coefficients from a multinomial logistic regression. Standard errors clustered bycountry in parentheses; ***p, 0.01, **p, 0.05, *p, 0.1.


difficult than predicting one of two in a binomial logit model. We correctly predict thepositions of all parties in two countries (Austria and Latvia); in only four countries(Czech Republic, Germany, Slovenia and Sweden) is the percentage of correct predictionsless than 50 percent (see Table 4).27

Figures 1a and 1b offer a graphic illustration of the substantive effects of two of ourvariables. The first of the two graphs shows the effect of timing on the probability of thethree outcomes. The most important finding from this graph is that the probability that aparty will not endorse a referendum (the bold line) follows an inverted-U shape, and theprobability falls sharply in the last two years before an election. While the likelihood thata party will choose to endorse a referendum and support the treaty remains relativelystable throughout the first few years of the electoral cycle, it declines in the last year. Bycontrast, the probability of a party campaigning against the treaty increases sharply rightbefore an election. The 95 percent confidence intervals (not shown) demonstrate thatthese changes across the electoral cycle are statistically significant. The confidence intervalfor the ‘‘not endorse’’ curve at its peak (900 days before the next election) ranges from0.43 to 0.91. By contrast, the confidence interval for the same curve right before anelection ranges from 0 to 0.01.The second of the two figures illustrates the effect of public opinion on party

positioning. Most obviously, the probability that a party will not endorse a referendumstrongly increases, together with public support for the treaty, from 0.03 [0,0.13] to 0.76[0.51,1.00]. This means that parties were more likely to endorse a referendum on theConstitutional Treaty in countries with a skeptical public opinion. By contrast, theprobability that a party will endorse a referendum but oppose the treaty falls from 0.88[0.74,1.00] to 0.02 [0,0.06] as public opinion becomes increasingly more favorable towardthe treaty. Parties endorsed a referendum and came out in favor of the treaty in countrieswith a medium degree of public support for the treaty.

TABLE 4 Number of Parties per Country and Number and Percentage CorrectlyPredicted

Countries Parties Correct % Countries Parties Correct %

Austria 4 4 100 Latvia 9 9 100Belgium 10 5 50 Lithuania 8 5 63Cyprus 7 5 71 Luxembourg 6 3 50Czech Republic 5 2 40 Malta 2 1 50Denmark 8 7 88 Netherlands 9 8 89Estonia 6 4 67 Poland 7 5 71Finland 7 5 71 Portugal 6 4 67France 8 7 88 Slovenia 5 2 40Germany 6 2 33 Slovakia 8 6 75Greece 4 2 50 Spain 12 10 83Hungary 4 3 75 Sweden 7 2 29Italy 12 11 92 United Kingdom 9 8 89Overall 169 120 71

27 We also used a ROC curve to assess the fit of the model. The Area Under Curve (AUC)value of 0.999 suggests that our model performs extremely well. Table 13 in the Appendix alsoshows the predictions and actual observations disaggregated across the three values of our dependentvariable.


Interaction Effects

Our model suggests that the effects of both Public Support and Political Capital shoulddepend on the timing of the referendum within the election cycle. We test theseexpectations by including interaction terms between these two variables and Timing in ourempirical model. The results of the model that includes the interaction effect with PublicSupport are very supportive (Model 2 in Table 3). Figure 2 displays the substantive effectof this interaction effect,28 showing that the probability of not endorsing a referendumincreases over time if public support for the treaty is high (in this case the maximum). Bycontrast, with low public support for the treaty (the minimum value in the dataset), theprobability of not endorsing a referendum falls very sharply around the middle of the

1500 1000 500 0

0.0

0.2

0.4

0.6

0.8

1.0

Timing (days left to next election)

Pro

babi

lity

Endorse, treaty yes (EY)Endorse, treaty no (EN)Not endorse (NE)

60 65 70 75 80 85 90 95

0.0

0.2

0.4

0.6

0.8

1.0

Public support

Pro

babi

lity

Endorse, treaty yes (EY)Endorse, treaty no (EN)Not endorse (NE)

Fig. 1. Predicted effects of Timing and Public Support

1500 1000 500 0

0.0

0.2

0.4

0.6

0.8

1.0

Timing (days left to next election)

Pro

babi

lity

(no

refe

rend

um)

Maximum supportMinimum support95% CI

Fig. 2. Predicted effect of Timing on the probability of no referendum endorsement in model with interaction term

28 For the models with interaction effects we center the interacted variables. The predicted effects arecalculated while keeping all other variables at their mean value or mode (for dichotomous variables).


electoral cycle. The interaction effect between Political Capital and Timing, by contrast, isnot statistically significant at the conventional level. It just misses that mark, however,and the term is weakly statistically significant when using a multinomial probit model (seeModel A3 in the Appendix).

Robustness Checks

We carried out a series of tests to check the robustness of our results (all of these resultsare shown in the Appendix). First, we dropped all 62 government parties from the model(Model A4 in the Appendix) because, according to our theoretical model, governmentparties only have two options available to them: to endorse or reject a referendum.29 Theirstance on the treaty itself is predetermined by the fact that they actively participated in itsnegotiation. Including them in the multinomial model with three options may thereforebias our results for the EY vs. EN comparison. In fact, the results of this model come evencloser to our expectation than those reported in Model 1, with the coefficient for Politicalcapital now no longer statistically significant in the EY vs. EN comparison.Secondly, we check whether the results are robust to dropping 57 parties with extreme

positions on European integration. We operationalize ‘‘extreme position’’ as having avalue beyond one standard deviation around the mean of Party Benefit (Veen 2011). Thereason for dropping these parties is the same as for government parties: realistically, theydo not have any flexibility to alter their position on the treaty itself, as their stance ispredetermined by their ideological orientation. Again the results are very stable, with theexception of the coefficient for Public Support, which is no longer statistically significant(Model A5 in the Appendix).Thirdly, we estimate a model with random intercepts to account for the multilevel

structure of our data more directly, namely parties nested within countries.30 Most of theresults from this model are similar to those from the model without random effects(Model A6 in the Appendix). The exception is Timing, which is no longer statisticallysignificant. However the timing of the decision about whether to call a referendum in theelectoral cycle still matters in this model, as the highly statistically significant coefficientfor Timing2 shows.Fourthly, we use different measures for some of our variables. On the one hand, we use

data from Hooghe et al. (2010) to operationalize Party Benefit. Doing so does not changethe results reported above (Model A7 in the Appendix). On the other hand, using analternative measure for Political Capital (approval of the government’s record to date)does not change most of the results, but the coefficient for that variable is only weaklystatistically significant in the EN vs. NE comparison (Model A8 in the Appendix).Fifthly, we add a measure of internal conflicts within parties, as it has been argued that

internally divided parties may be more likely to call for a referendum (Bjørklund 1982).While ideally we would have data on internal divisions on the Constitutional Treaty itself,the best data that we could find is the variable internal dissent on European integration

29 Our theoretical model, however, leads to the expectation that domestic political factors do not exerta differential effect on the position of government and opposition parties on whether or not to call areferendum, which explains why we do not drop government parties from our main model. Incumbencyonly matters with respect to the support for the treaty itself, given that opposition parties enjoy a lot moreleeway in that respect. This also becomes clear in the comparative statics results presented in the onlineappendix.

30 For this model we use standardized variables, as recommended by Gelman and Hill (2007).


from the 2002 Chapel Hill Expert Survey (Hooghe et al. 2010). Alas, we are missing valuesfor this variable for 35 parties. Nevertheless, the results are very robust to the inclusion ofthis variable (Model A9 in the Appendix). Finally, we drop Germany from our analysis, asits constitution makes the initiation of a referendum on an EU treaty particularly difficult.31

Again, doing so does not affect our results (Model A10 in the Appendix).

CONCLUSION

We have presented a game-theoretic model of the process of ratifying an international treatyin order to explain parties’ simultaneous decisions about whether to call for a referendum andwhat position to take on the treaty itself. Our findings show that this decision depends on thetiming of the decision within the electoral cycle, the incumbent’s political capital and the levelof public support for the treaty. The higher the government’s relative political capital, the lesslikely parties are to endorse a referendum and reject the treaty. The probability that partieswill call for a referendum is lowest in the middle of the electoral cycle. Finally, public supportfor the treaty reduces the probability that parties will opt for a referendum. These findings arerobust to variations in data sources and estimation techniques. What makes our approachparticularly powerful is that for given referendum-initiation rules, our findings on party-levelstrategies are a sufficient statistic for predicting country-level outcomes.The results have implications for a variety of debates. First, our findings run counter to

fears that parties may use popular votes mainly as plebiscites. On the contrary,uncertainty and concerns about second-order voting contribute to a situation in whichparties are most likely to call for a referendum on an international treaty when publicopinion is rather skeptical. This finding offers an explanation for the sizable number offailed referendums on European integration.The results also speak to the literature on second-order elections. For some time, scholars

have debated the extent to which voters use referendums to send signals to theirgovernment.32 Our findings add to this debate by indicating that parties base their decisionon whether to call for a referendum on the assumption that voters engage in second-ordervoting. The fear of being punished in a referendum makes government parties wary ofcalling for a referendum even if voters are satisfied with the treaty in question.Finally, for the specific case of European integration, the article’s findings suggest that

making the decision about whether to call for a referendum on treaty revisions (or issuessuch as the accession of Turkey) at the national level is problematic. With respect to theratification of EU treaties, the fact that domestic political factors influence the incidenceand outcome of a referendum makes it more difficult to defend the application of directdemocracy at the national level on normative grounds. It will always be difficult toinsulate the international policy debate and decision-making process from domestic-levelfactors and contingencies, which may end up distorting popular preferences with respectto integration policy, obfuscating the interpretation of the referendum outcome orderailing the overall ratification process. A European-level referendum on major decisionsfacing the EU may be the best instrument to deal with this problem (Rose 2011).

31 To be precise, a referendum on an international treaty could only be called after amendingGermany’s Basic Law with a two-thirds majority. In the wake of the debate on the Constitutional Treaty,such a change was indeed proposed, explaining why we do not drop the country from our main model.Table 8 in the Appendix offers more information on constitutional provisions.

32 See, for example, Franklin (2002), Svensson (2002) and Garry, Marsh and Sinnott (2005).


REFERENCES

Ashworth, Scott, and Ethan Bueno de Mesquita. 2006. ‘Monotone Comparative Statics for Models

of Politics’. American Journal of Political Science 50(1):214–31.

——. 2009. ‘Elections with Platform and Valence Competition’. Games and Economic Behavior

67(1):191–216.

Bjørklund, Tor. 1982. ‘The Demand for Referendum: When Does It Arise, When Does it Succeed?’

Scandinavian Political Studies 5(3):237–59.

Braun, Daniela, Maike Salzwedel, Christian Stumpf, and Andreas Wust. 2007. ‘Euromanifesto

Documentation’. Available at http://www.ees-homepage.net

Carrubba, Clifford J. 2001. ‘The Electoral Connection in European Union Politics’. Journal of

Politics 63(1):141–58.

Closa, Carlos. 2007. ‘Why Convene Referendums? Explaining Choices in EU Constitutional

Politics’. Journal of European Public Policy 14(8):1311–32.

de Vries, Catherine E. 2007. ‘Sleeping Giant: Fact or Fairytale? How European Integration Affects

National Elections’. European Union Politics 8(3):363–85.

——. 2009. ‘The Impact of EU Referenda on National Electoral Politics: The Dutch Case’.

West European Politics 32(1):142–71.

Dow, Jay K., and James W. Endersby. 2004. ‘Multinomial Probit and Multinomial Logit: a

Comparison of Choice Models for Voting Research’. Electoral Studies 23:107–22.

Dur, Andreas, and GemmaMateo. 2011. ‘To Call or Not to Call: Political Parties and Referendums

on the EU’s Constitutional Treaty’. Comparative Political Studies 44(4):468–92.

Eurobarometer. 2003. ‘Convention on the Future of Europe – Wave 2’. Flash Eurobarometer 142(2).

——. 2003b. ‘Intergovernmental Conference’. Flash Eurobarometer 152.

——. 2004a. ‘Future European Constitution’. Flash Eurobarometer 159.

——. 2004b. ‘Public Opinion in the European Union’. Standard Eurobarometer 61.

——. 2005. ‘The Future Constitutional Treaty’. Special Eurobarometer 62(1).

European Election Study. 2009. Design, Data Description and Documentation, 2nd edition.

Mannheim, Germany: Mannheimer Zentrum fur Europaische Sozialforschung.

Finke, Daniel, and Thomas Konig. 2009. ‘Why Risk Popular Ratification Failure? A Comparative

Analysis of the Choice of the Ratification Instrument in the 25 Member States of the EU’.

Constitutional Political Economy 20(3/4):341–65.

Franklin, Mark N. 2002. ‘Learning from the Danish Case: A Comment on Palle Svensson’s Critique

of the Franklin Thesis’. European Journal of Political Research 41(6):751–57.

Gabel, Matthew, and Kenneth Scheve. 2007. ‘Estimating the Effect of Elite Communications

on Public Opinion Using Instrumental Variables’. American Journal of Political Science

51:1013–28.

Gallagher, Michael, and Paul Mitchell, eds. 2008. The Politics of Electoral Systems. Oxford and

New York: Oxford University Press.

Garry, John, Michael Marsh, and Richard Sinnott. 2005. ‘ ‘‘Second-order’’ versus ‘‘Issue-voting’’

Effects in EU Referendums: Evidence from the Irish Nice Treaty Referendums’. European

Union Politics 6(2):201–21.

Gelman, Andrew, and Christopher Hill. 2007. Data Analysis Using Regression and Multilevel/

Hierarchical Models. Cambridge, UK: Cambridge University Press.

Hobolt, Sara B. 2006. ‘How Parties Affect Vote Choice in European Integration Referendums’.

Party Politics 12(5):623–47.

——. 2009. Europe in Question: Referendums on European Integration. Oxford: Oxford University

Press.

Hooghe, Liesbet, Ryan Bakker, Anna Brigevich, Catherine de Vries, Erica Edwards, Gary Marks,

Jan Rovny, Marco Steenbergen, and Milada Vachudova. 2010. ‘Reliability and Validity of

the 2002 and 2006 Chapel Hill Expert Surveys on Party Positioning’. European Journal of

Political Research 49(5):687–703.


Hug, Simon. 2004. ‘Occurrence and Policy Consequences of Referendums: a Theoretical Model and

Empirical Evidence’. Journal of Theoretical Politics 16(3):321–56.

Hug, Simon, and Tobias Schulz. 2007. ‘Referendums in the EU’s Constitution Building Process’.

Review of International Organizations 2(2):177–218.

Long, Scott J., and Jeremy Freese. 2006. Regression Models for Categorical Dependent Variables

Using Stata, 2nd edition. College Station, TX: Stata Press.

Marsh, Michael, and Slava Mikhaylov. 2010. ‘The European Parliament Elections as Second-order

National Elections: A Review of Evidence’. Living Reviews in European Governance 5(1).

Available at http://europeangovernance.livingreviews.org/open?pubNo5lreg-2010-4&page5

articlese4.html

Putnam, Robert D. 1988. ‘Diplomacy and Domestic Politics: The Logic of Two-level Games’.

International Organization 42(3):427–60.

Ray, Leonard. 2003a. ‘Reconsidering the Link between Incumbent Support and pro-EU Opinion’.

European Union Politics 4(3):259–79.

——. 2003b. ‘When Parties Matter: the Conditional Influence of Party Positions on Voter Opinions

about European Integration’. The Journal of Politics 65(4):978–94.

Reif, Karlheinz, and Hermann Schmitt. 1980. ‘Nine Second-order National Elections: A Conceptual

Framework for the Analysis of European Election Results’. European Journal of Political

Research 8(1):3–44.

Rose, Richard. 2011. ‘The Case for Pan-European Referenda’. Europe’s World available at http://

www.europesworld.org (accessed 7 August 2013).

Sattler, Thomas, and Johannes Urpelainen. 2012. ‘Explaining Public Support for International

Integration: How Do National Conditions and Treaty Characteristics Interact with Indivi-

dual Beliefs?’ Journal of Politics 74(4):1108–24.

Schmitt, Hermann. 2005. ‘The European Parliament Elections of June 2004: Still Second-order?’

West European Politics 28(3):650–79.

Schneider, Gerald., and Patricia A. Weitsman. 1996. ‘The Punishment Trap: Integration Referendums

as Popularity Contests’. Comparative Political Studies 28(4):582–607.

Schofield, Norman. 2005. ‘A Valence Model of Political Competition in Britain: 1992–1997’.

Electoral Studies 24(3):347–70.

Setala, Maija. 1999. ‘Referendums in Western Europe: A Wave of Direct Democracy?’ Scandinavian

Political Studies 22(4):327–40.

Steenbergen, Marco, Erica Edwards, and Catherine de Vries. 2007. ‘Who Is Cueing Whom?

Mass-elite Linkages and the Future of European Integration’. European Union Politics

8(1):13–35.

Svensson, Palle. 2002. ‘Five Danish Referendums on the European Community and European

Union: A Critical Assessment of the Franklin Thesis’. European Journal of Political Research

41(6):733–50.

Tillman, Erik R. 2004. ‘The European Union at the Ballot Box? European Integration and Voting

Behavior in the New Member States’. Comparative Political Studies 37(5):590–610.

Trechsel, Alexander. 2010. ‘Reflexive Accountability and Direct Democracy’. West European

Politics 33(5):1050–64.

Tridimas, George. 2007. ‘Ratification through Referendum or Parliamentary Vote: When to Call a

Non-required Referendum?’ European Journal of Political Economy 23(3):674–92.

Veen, Tim. 2011. ‘Positions and Salience in European Union Politics: Estimation and Validation of

a New Data Set’. European Union Politics 12(2):267–88.


Documents

International Treaty Ratification and Party Competition: Theory and Evidence from the EU's Constitutional Treaty