Regionalism, Foreign Policy and Executive/Legislative Relations: Mercosur and the European Union in Comparative Perspective

Regionalismo, Política Externa e Relações Executivo/Legislativo: O Mercosul e a União Europeia em Perspectiva Comparada

Régionalisme, Politique Étrangère et Relations Exécutif/Législatif: Le Mercosur et l’Union Européenne en Perspective Comparée

Regionalismo, Política Exterior y Relaciones Ejecutivo/Legislativo: El Mercosur y la Unión Europea en Perspectiva Comparada

Abstract

In a globalized world, Legislative control over matters related to Mercosur and the European Union (EU) is important. However, its importance varies according to Regional Foreign Policy (RFP) decisions. This article investigates whether there are configurations, considering institutional design, political conditions, and economic factors, which explain the intensity of legislative participation in determining such matters. Specialists were consulted to select five decisions that represented significant changes for the regional dynamics of Mercosur and EU member countries. After defining the intentional sample of cases, necessary and sufficient conditions for the participation of the Legislature in RFP were identified through a Qualitative Comparative Analysis (QCA). The results indicate that institutional design and economic factors matter, but prior debates in Parliaments on the subject are also relevant to explaining legislative action. This was corroborated by a Coincidence Analysis (CNA) that demonstrated that this variable makes up the causal chain of participation.

regional foreign policy; qualitative comparative analysis; legislative; Mercosur; European Union

Resumo

Em um mundo globalizado, o controle do Legislativo sobre matérias ligadas ao Mercosul e à União Europeia importa, mas varia em relação às decisões de Política Externa Regional (PER). Este artigo investiga se existem configurações, considerando o desenho institucional, as condições políticas e os fatores econômicos, que expliquem a intensidade da participação legislativa na determinação de tais matérias. Foi realizada uma consulta com especialistas para selecionar 5 decisões que representaram mudanças significativas para a dinâmica regional dos países mercosulinos e europeus. Após definida a amostra intencional de casos, através de uma Análise Qualitativa Comparativa (QCA), foram identificadas condições causais, necessárias e suficientes, para a participação do Legislativo em PER. Os resultados indicam que desenho institucional e fatores econômicos importam, mas debates prévios nos Parlamentos sobre o tema também são relevantes para explicar a atuação legislativa. Isso foi corroborado por uma Análise de Coincidência que demonstrou que essa variável compõe a cadeia causal da participação.

política externa regional; análise qualitativa comparativa; legislativo; Mercosul; União Europeia

Résumé

Dans un monde globalisé, le contrôle du législatif sur les questions liées au Mercosur et à l’Union Européenne est important, mais il varie en fonction des décisions de Politique Étrangère Régionale (PER). Cet article examine s’il existe des configurations, en tenant compte de la conception institutionnelle, des conditions politiques et des facteurs économiques, qui expliquent l’intensité de la participation législative dans la détermination de telles questions. Une consultation d’experts a été réalisée pour sélectionner 5 décisions représentant des changements significatifs pour la dynamique régionale des pays du Mercosur et de l’Union européenne. Après avoir défini l’échantillon intentionnel de cas, grâce à une Analyse Qualitative Comparative (QCA), des conditions causales, nécessaires et suffisantes, ont été identifiées pour la participation du législatif à la PER. Les résultats indiquent que la conception institutionnelle et les facteurs économiques importent, mais les débats préalables dans les Parlements sur le sujet sont également pertinents pour expliquer l’action législative. Cela a été corroboré par une Analyse de Coïncidence qui a démontré que cette variable fait partie de la chaîne causale de la participation.

politique étrangère régionale; analyse qualitative comparative; législatif; Mercosur; Union Européenne

Resumen

En un mundo globalizado, el control legislativo sobre los asuntos relacionados con el Mercosur y la Unión Europea es importante, pero varía en relación con las decisiones de Política Exterior Regional (PER). Este artículo investiga si existen configuraciones, teniendo en cuenta el diseño institucional, las condiciones políticas y los factores económicos, que expliquen la intensidad de la participación legislativa en la determinación de tales asuntos. Se consultó a expertos para seleccionar cinco decisiones que representaban cambios significativos en la dinámica regional del Mercosur y de los países europeos. Una vez definida la muestra intencional de casos, se utilizó un Análisis Cualitativo Comparativo (ACC) para identificar las condiciones causales necesarias y suficientes para la participación de la legislatura en PER. Los resultados indican que el diseño institucional y los factores económicos importan, pero los debates previos en los parlamentos sobre el tema también son relevantes para explicar la acción legislativa. Esto fue corroborado por un Análisis de Coincidencias que demostró que esta variable conforma la cadena causal de la participación.

política exterior regional; análisis cualitativo comparativo; legislativo; Mercosur; Unión Europea

Introduction

What conditions determine the level of participation of national Legislatures? Given the importance attributed to regional spaces - mainly from the 1980s onwards when notions of globalization and regionalism gained strength - States began to look not only at internal demands, that are linked to social and democratic issues but also – and increasingly – at external ones to maximize economic effectiveness.

This article examines whether the Foreign Policies of the States that make up the Southern Common Market (Mercosur) and the European Union (EU) are subject to the control of their Legislative Branches. It considers that the involvement of these institutions at the international level legitimizes the commitments and strengthens the democratic mechanisms of checks and balances. Specifically, this paper aims to find which configurations out of institutional design, political conditions, and economic factors lead to increased participation of legislators in Regional Foreign Policy (RFP).

Based on consultation with specialists, five cases of significant change in participation within national legislatures were selected in Mercosur and the EU: (a) the Treaty of Asunción; (b) the Adhesion of Venezuela to Mercosur; (c) the Ushuaia Protocol; (d) the Treaty of Maastricht; and (e) the Treaty of Lisbon.

Using the recommendations of experts from different fields and regions offers numerous advantages that enhance the study’s originality, minimize biases, and boost its reliability. The consultation of multiple experts can yield a study that is more equitable and impartial in comparison to a case selection solely conducted by authors.

We were able to identify instances for qualitative comparative analysis (QCA) looking to point out the causal conditions for the participation of the Legislature in Regional Foreign Policy. The operationalized model postulates that institutional design and political and economic conditions are relevant to legislative activity in RFP matters. This finding was later corroborated by a Coincidence Analysis (CNA), a method that considers causal chains between an outcome and all factors included in the model, a feature that is not present in QCA. It made it possible to identify causal sequences among variables and the desired outcome, enabling the CNA to confirm that prior debate in Parliaments is an important component of the causal chain of participation.

The literature on the relationship between the Legislature and Foreign Policy, especially the factors influencing the duration, intensity of debate, and veto power, continues to evolve. While there is a more extensive body of work available in some European countries, Latin America and Brazil, in particular, still lack sufficient scholarly exploration in this area. Additionally, the absence of systematically organized official data on legislatures posed challenges to our analysis. Nevertheless, this article introduces innovations to address this subject, shedding light on the pivotal role of legislative moderation in legitimizing RFP. It also underscores the considerable influence of institutional and economic factors on legislative engagement in RFP. In doing so, this article makes a valuable contribution to the ongoing debate and identifies promising avenues for future research.

In the first section, we discuss concepts related to the regionalism process and highlight the cases of the EU and Mercosur. In the second, we focus on the role of domestic conditions in the development of States’ RFPs. In the third section, we discuss our research design and the cases selected. In the fourth section, we present the results and, in the last, our conclusions.

Regional Foreign Policy

Sometimes interpreted as contradictory and sometimes as complementary, the notions of globalization and regionalism are at the center of a debate whose background is related to the challenging issue that reform poses to the State. It consists of the States’ necessity to conform to internal demands in the search for economic efficiency, and to external ones that are linked to social and democratic issues. Regionalization, thus, emerges as an intermediary space in which States can initiate reforms and engage in liberal economic practices before facing fully-fledged globalization.¹

The process of building an integrated zone naturally leads to questioning some of the concepts that have always been dear to the values of the nation-state. Among these values, the most salient one seems to be sovereignty. The nation-state resists relinquishing its prerogatives in favor of other instances, be they supra or infra-state. So, the State finds itself in a paradoxical situation, where the international level encourages reforms that could threaten its sovereign nature. Within this context, not only does the concept of sovereignty change, but the ideas of legitimacy, representation, and decision-making processes also change – the latter two translated into structures that underly the former (Fonseca, 1998).

Given these considerations, it is important to highlight that the complexity of the EU system is due not only to its multiple arrangements but also to its devices of checks and balances designed to submit those arrangements to a system of control. Since its establishment via the Maastricht Treaty, the basic themes of the EU have been defined: European Communities, Common Foreign and Security Policy, Justice, and Internal Affairs.

Formerly, with only consultive powers, the EP had a historical process of empowerment and development of Legislative functions which were targeting the coordination of national and regional policies (Lessa, 2003). The Legislative and Control Branches of the European Parliament were strengthened with the introduction of the co-decision procedure, the expansion of the cooperation process,² and, the need for its approval for some political and institutional issues (such as enlargement for new members, association with other countries, and international agreements) (Lessa, 2003).

In 2007, employing Art. 14 of the Treaty on European Union (TEU) 1, the Treaty of Lisbon transformed the European Parliament (EP) into a co-legislator of the EU together with the Council. And, through Art. 5 of the TEU, invited national Parliaments to ensure respect for the principle of subsidiarity and to be alert for Commission proposals without a legal basis (Art. 352 of the Treaty on the Functioning of the European Union). In other words, legislative control is performed at two distinct levels, the national and the supranational.

Thus, the EU - and specifically the EP - is the most sophisticated experience of regional integration and it still influences many institutional processes worldwide, including the Southern Common Market (Mercosur) (Medeiros et al., 2015).

In the case of Mercosur, although mechanisms for harmonizing laws and formulating policies have been provided since the Treaty of Asunción (1991), it was only in 1994 with the Ouro Preto Protocol that regional legislative representation was established in the form of the Joint Parliamentary Commission (“CPC”, Portuguese acronym). Despite playing an important role in establishing links between the regional and national legislative spheres, the CPC mainly performed consultative functions. Later, “with the deepening of the integration process, the need for greater participation of the National Legislatures became clear” (Mercosur Parliament, 2021). Thus, in 2006, the Constitutive Protocol of the Mercosur Parliament was approved and gave birth to “Parlasur,” the Mercosur Parliament. This institutional reform sought the establishment of a link between democratic and representative topics, guaranteeing greater dialogues with the society and Legislative participation in RFP (Medeiros, Amelotti, Moura, 2021).

Despite its symbolic value and the fact that this is the most developed case of a regional parliament in Latin America (Mariano et al., 2017), the Parlasur, just as with the CPC, is limited to an eminently consultative role vis-à-vis executive institutions. Hence the bloc’s legislative control continues to be conducted within the domestic parliaments.

The incorporation of institutionalized legislative participation in the bargaining process enables democratic nations to establish trustworthy commitments with other states, thereby expanding the realm of mutual confidence. On the other hand, legislative responses to executives who believe that legislators lack a legitimate role in foreign policy will cast doubts on the credibility of democratic nations’ commitments (Martin, 2000). Even so, the conditions and the probability of international cooperation are contingent upon the power dynamics between the legislative and executive branches (Milner, 1997).

Put simply, it is anticipated that robust democracies rely on the involvement of the Legislative branch in both domestic and foreign decision-making processes to ensure the credibility of their commitments. However, the extent of this participation varies based on the internal dynamics specific to each nation. Within the framework of regionalism, this article is specifically interested in scrutinizing how national legislatures intervene in the decision-making process of States in the Mercosur and EU.

If such moderation exists, it will reinforce the arguments of Moravcsik (2002), Majone (1998) and Scharpf (2000). These authors attempt to move away from the State-centric paradigm introducing nuances that relativize the democratic deficit. They especially consider advanced regional integration processes such as the EU as they seek to privilege new channels of legitimation. For instance, Moravcsik (2002) argues that the concern with the EU’s democratic deficit is misplaced, as it is judged against advanced industrial democracies rather than as an ideal and legitimate plebiscitary or parliamentary democracy.

By contrast, if we find that there exists no such moderation by the legislature, this will strengthen the criticisms of Follesdal and Hix (2008), Dahl (1999), Rubenfeld (2004), and Rabkin (2005). Dahl (1999) argues that the EU’s bureaucratic nature, its disjunction between national and supranational spheres, and its lack of citizen participation undermine its accountability and deliberation devices; Rubenfeld (2004) emphasizes the markedly bureaucratic, diplomatic, and technocratic nature of international organizations, which he argues are therefore far from public control; In turn, Rabkin (2005) argues that the dominant role of elites in establishing the agendas of multilateral institutions makes them unresponsive to citizen demands.

The fact is that decision-making in Foreign Policy can be influenced by such factors as public opinion, the nature of the political regime, the form of government, and the national economic scenario. These elements act decisively and not peripherally.

The next section presents some of the factors that influence the behavior of States regarding their Regional Foreign Policies.

Domestic determinants of Regional Foreign Policy

Researchers can improve their explanatory and predictive power regarding international relations by looking at leaders’ interests and the domestic conditions under which they operate. In other words, international relations are not a kind of high politics separate from domestic politics (Bueno de Mesquita, Smith, 2012). Self-interested leaders make Foreign Policy choices and political institutions systematically shape these choices and their impact on the globalized world.

However, representation and decision-making processes in matters of Foreign Policy are historically assigned to the Executive Branch. According to the classic debate proposed by Almond (1956), international relations are characterized by low participation of domestic actors other than the Executive, except concerning issues considered important by public opinion. This notion is further reinforced and elaborated by Scott (1999), who demonstrates that, when confronted with unpopular policies, Congress becomes more proactive and challenges Executive decisions. Therefore, public opinion plays a significant role in influencing congressional activism and the degree of rivalry between branches in foreign policymaking (Scott, 1999).

This is consistent with Putnam’s (1993) assertion that international negotiations are conceived as a two-level game that simultaneously occurs within national and international arenas through complex networks of interaction. In short, at the national sphere, groups pursue their interests by pressing the government to adopt policies that are favorable to them – and here Parliament can play a crystalizing role. At the international level, these same governments seek to maximize their capacity to satisfy domestic pressures, while minimizing the adverse consequences this can entail for foreign partners (Putnam, 1993).

In this same rationale, Milner (1997) also presents contributions to understanding how closely linked are the domestic and international arenas and how their interactions shape the cooperation between States. According to her, variations in the preferences and distribution of information among domestic actors affect the game of internal ratification and international cooperation. Milner (1997) concludes that no political leader can afford to ignore domestic politics when contemplating Foreign Policy choices.

The role of the national Legislature in Foreign Policy

According to Pinheiro (2008), themes related to Foreign Policy are not restricted to the Executive Branch and the Legislature is a relevant actor capable of intervening in the process of forming trade policy. This argument is in line with Almond’s thesis (1956), as parliamentary intervention is prompted by the relevance of trade to the salient theme of the economy. Almond (1956) advocates that economic variables and public opinion are fundamental factors in raising the awareness of legislative bodies, which are periodically subject to electoral scrutiny. Within this perspective and according to Santos et al. (2020), economic growth (GDP), market concentration (HHI), and commercial opening do matter concerning the interests of legislative bodies in participating in foreign policy.

However, the Legislature enjoys specific prerogatives in respect of Foreign Policy. For example, during the process of approving an international agreement, when its ratification is the task of the Parliament (ex post action). Or in secondary phases such as when it must approve changes in domestic legislation or a specific budget allocation. Another form of action is the establishment of congressional monitoring mechanisms (Pinheiro, 2008).

In a study on the question of the international credibility of States, Martin (2000: 202) points out that the rigor of a government’s commitments is the focal point of its external reciprocal relations. She argues that this type of credibility is related to an institutionalized role of the Legislative Branch at the domestic level, which can influence the international cooperation processes.

To support her argument, Martin (2002) empirically analyzes the case of the EU and concludes that, even if all member states of the bloc have some level of parliamentary democracy, there is significant variation in terms of the degree of influence that the Parliament has on the activities of ministers at the EU level.

Parliaments are potentially destructive to the course of economic integration, which means that States that negotiate without parliamentary barriers have the best cooperation ties with the EU (Martin, 2002). This logic leads us to expect that States that negotiate without parliamentary restrictions will have the best records of cooperation within the EU. According to Scott (1999) and Martin (2002), in case of convergence of interests, the Legislature delegates; in case of divergence, it seeks to increase its participation.

However, empirical evidence has shown that the State with the most institutionalized parliamentary oversight of EU negotiations - Denmark - has consistently the best record of implementing EU legislation (Martin, 2002). Overall, the degree to which the Parliament constrains government ministers is strongly and positively correlated with implementation rates. This reinforces the idea that Parliament is an inseparable part of formulating Foreign Policy, at least in the European context.

Moreover, Carter and Scott (2012) demonstrate that legislative behavior in foreign policy can encompass activism, assertiveness, and entrepreneurship. The first pertains to influencing decisions, whether in favor or opposition, the second involves challenging the Executive, while the third relates to individual legislative attitudes. In addition, the combination of the institutional dimensions – activism and assertiveness – gives rise to four models: competition, disengagement, support, and strategy. Both competitive and supportive attitudes imply higher levels of engagement, while strategic behavior is associated with activism on specific issues. Disengagement corresponds to inactivity and acquiescence (Carter, Scott, 2012).

In this way, the nature of the Foreign Policy emerges as one of the core elements of the Executive-Legislature relationship in matters of negotiations with foreign entities. This means that institutional design, which regulates Executive-Legislature relations, is important in shaping the RFP. However, institutions do not operate in a vacuum. They are mediated by domestic institutional design and political relations and are strongly influenced by the economic scenario. As previously stated, this article intends to verify how these variables are combined, determining the role that the Legislature has in the formulation of Foreign Policy.

Domestic institutional design

The external/internal dichotomy has guided political thought for many centuries. Hannah Arendt (1998) states that the Roman origin of the concept of Foreign Policy is unquestionable since it was the Roman politicization of space that defined the Western world. This indicates that, despite being of an external nature, international politics follows the conventional democratic procedures of internal politics.

It is a common prerogative that the participation of the Legislature in parliamentary systems tends to be more intense than in other systems because the government’s survival depends on the support of a parliamentary majority. In the United Kingdom, for example, although it has a strong Executive, the prime minister’s (PM) need for a majority in the House of Commons can fatally weaken its position (Lijphart, 2012). Which, in turn, threatens the Executive’s power to make decisions and the very continuity of government.

Similarly to parliamentary democracies, where the ratification of Treaties demands legislative approval by vote or popular Referendums, presidential systems require presidents to submit Treaties for consideration by Congress and to secure majority vote support (Lantis, 2006). Nonetheless, in majoritarian parliamentary systems, PMs hold an institutional advantage in garnering support for favored policies (Lantis, 2006). Whereas in presidential systems, the Legislature often employs institutional mechanisms to oversee and regulate presidential actions in the international arena (Feliú, 2018). Taking Brazil as an example, the Legislature “relies on a set of institutionalized instruments that allow it to intervene when it feels so inclined ex ante and ex post in the production, execution, and monitoring of Brazilian Foreign Policy” (Anastasia; Mendonça, Almeida, 2012:632). Furthermore, these instruments are used effectively and there is a complex network of interactions between the actors involved in the decision-making process.

However, unlike the Brazilian case where the Legislature directly influences foreign policy decisions, Uruguay employs the practice of Presidential Diplomacy or Summit Diplomacy (Preto, 2006). That is, the President of the Republic directly participates in decision-making processes and international summits (Medeiros et al., 2021). It is worth noting that even with the difference from Brazil’s example, Uruguayan political parties influence legislative processes related to international matters and act as veto players in the International Affairs Commission (Tsebelis, 2022 apud Medeiros et al., 2021).

Looking at the Latin American case, recent debates have given much attention to the effective role played by Parliaments in countries’ decision-making processes. Studies have revealed a heterogeneity of Legislatures in Latin America resulting from the wide variety of constitutional powers delegated to Presidents, the party system, and the legislature itself (Pinheiro, 2008). This can take place around one or more assemblies and interact within federal or unitary systems, in large, medium-sized, and small countries (Llanos, Nolte, 2003). Therefore, and aligned to Lantis (2006), the regime type emerges as an important condition for understanding the ratification process of international agreements.

In addition, the literature points to another component of institutional design that can be a determinant of foreign policy results: whether the legislature is uni- or bicameral. Unicameral system designs tend to involve fewer actors with veto power, which, in theory, gives the Executive greater autonomy. Bicameral structures open up a new decision-making arena characterized by two institutional actors who can actively participate in the decision-making process.

Nonetheless, bicameral systems have their own peculiarities, which can mean a strong system or a weak one (Lijphart, 2012). This variation arises due to each chamber’s unique configuration and influences the decision-making process. Moreover, standing committees and subcommittees within each chamber play an important role in shaping and often formulating policies (Carter, Scott, 2012). The existence of a formal balance between the constitutional powers of each house (as in Argentina, Uruguay, and Italy) and the method of selection of the upper house (e.g., by direct elections such as in the above-mentioned countries) highlight the question of the democratic legitimacy of the two houses in a bicameral (Lijphart, 2012). This is an important aspect in determining the level of symmetry between them (Lijphart, 2012).

It appears to be the case that institutional design matters, which leads to an expectation that the formulation of Foreign Policy can count, to a greater or lesser extent, on the participation of the Legislature.

Political relations

In addition to checks and balances between the branches of government, democracy provides for the representation and participation of multiple actors. Regarding the more specific case of Foreign Policy, the influence of parties and the existence of fragmentation in the legislative houses are disputed subjects in the literature.

Taking Brazil as an example, Schneider (1976 apud Onuki, Oliveira, 2010) argues that parties do not matter when it comes to decision-making at the international level. Corroborating this perspective in writing on the Latin American case, Lima and Santos (2001:121) state that Foreign Policy is “the natural object of delegation of decision-making power from the Legislature to the Executive.” This is regarded as being due to the extensive nature of presidential powers, in addition to political parties’ low expertise in the area and the weak electoral value of foreign policy to them (Lima, Santos, 2001; Santos, 2006 apud Onuki, Feliú, Oliveira, 2009).

Onuki and Oliveira (2010:145) take a contrasting view and argue that “Brazilian political parties not only take distinct positions in respect of Foreign Policy but, depending on the structure of the government’s support base, they can also alter it.” Ribeiro and Pinheiro (2016) add to this argument saying that Legislative support for Latin American Presidents’ Foreign Policies may reflect variables such as coalition size, ideology, and the effective number of parties (hereinafter “ENP”).

Furthermore, in a country with a bicameral system, support for the Executive may also be determined by the “difference between the number of members of each house, which in turn influences the degree of party fragmentation” (Ribeiro, Oliveira, 2018:79). Thus, there is an expectation that the chamber with the largest number of members will present more difficulties in the formation of decision-making majorities (Ribeiro, Oliveira, 2018).

Writing about Latin American presidential systems, Ribeiro and Pinheiro (2016:474) state that “the high level of ideological dissociation among political parties in legislatures may result in weaker control over the legislative agenda by the Executive.” And about parliamentary systems, the number of parties in the political game “affects the ease with which governments can be formed” (Wolinetz, 2006:51). Thus, in political analysis it is useful to access the degrees of divergence inside the governments.

The effective number of parties³ is an indicator of party fragmentation that is widespread in the literature and can provide clues to the attitudes of the Executive and the Legislature to Foreign Policy, both in presidential and parliamentary systems. The ENP represents a weighted count of the parties that make up the political arenas and plays an important role in understanding decisions, as it affects the support given to the government leadership. The higher the ENP, the more difficult it is for the Executive to form a majority coalition and the greater the tendency towards ideological polarization within the government (Ribeiro, Pinheiro, 2016).

Writing about a political scenario dominated by many parties of similar sizes, Melo (2019) argues that the increased capacity for individual bargaining with the Executive, which this scenario gives rise to, increases the cost of coordinating the government support base.

The parties within the houses of the Legislature become more relevant to the decision-making process, making it more costly or fluid. Although there is no consensus on whether these effects are valid at the systemic level, some studies have provided strong evidence to support the idea that partisan issues explain Foreign Policy to some extent (Scott, 1999; Martin, 2002; Lantis, 2006; Pinheiro, 2008; Carter, Scott, 2012).

Clearly, a large ENP can hinder the Executive’s decision-making capacity. Our specific interest is in whether party fragmentation represents an opportunity for the Legislature, in such a way as to result in more active participation in RFP formulation.

The development of countries’ international relations can be explained in other ways and, as already discussed, institutional design and political relations appear to play a significant role. Other factors may also be relevant, such as the economy. Therefore, in the following section, we will present arguments about its impact.

The economic scenario

Economic policy and foreign policy exhibit different patterns in their relationship to overall government approval, as the former has a consistent impact over time, while the effect of the latter is variable (McAvoy, 2015). The public’s assessment of the President as seen through the prism of foreign policy and economic management suggests three potential answers: stability, change, or stability in one context and change in another.

Economic stability is one of the criteria for the electorate’s assessment of a President. Many studies have considered whether Presidents are assessed based on their past economic performance or future expectations about the economy. A finding common to almost all these studies is that the relationship between economic variables and presidential approval has not changed over time (McAvoy, 2015).

Therefore, economic and foreign policy are both important in the public’s assessment of the President (McAvoy, 2015). Furthermore, voters learn over time and change the way they weigh up foreign policy in their assessment of the President. Then, it is to be expected that economic development and Foreign Policy are relevant factors to explain the intensity of the participation of the Parliament in RFP.

Another important aspect is related to the preferences of the private sector. In scenarios of divergence among intentions of the Executive branch and the entrepreneurs about economic openness, the Legislative is, indirectly, called up to balance the Executive actions (Oliveira, Onuki, 2007). In the Brazilian case, for example, there is little parliamentary action in commercial matters given that the control of this agenda lies on the Executive branch (Oliveira, Onuki, 2007).

Even so, despite the impossibility of formulating theoretically oriented hypotheses concerning the relationship between economic performance and legislative participation, speculation is possible. It is feasible that, in a context where the government has achieved macroeconomic success, the Legislature withdraws from or reduces its participation in Foreign Policy formulation. Correspondingly, instability can motivate the Legislature to actively participate in Foreign Policy formulation as an alternative to dealing with economic problems.

In addition to performance, other structural aspects of the economy can affect the Legislature’s involvement in Foreign Policy. Santos, Cimini, and Bohigues (2020) found that this was the case in their analysis of the perceptions of parliamentary elites from seventeen Latin American countries vis-à-vis the economy and international relations. They sought to identify the determinants of parliamentarians’ behavior regarding political and economic relations between States in the region and between Latin American and non-Latin American States. According to them:

There are at least two aspects that are relevant to the debate on the participation of Parliaments and parliamentarians in political and economic international relations in Latin American countries. The first concerns the importance of games at two levels, domestic and international, and of the legislature in matters of a distributive nature – international trade. The second refers to the influence of economic elites in the legislature, observing the ways in which the demands of big business interests (industrial, agribusiness and finance) are reflected in Parliament. (Santos, Cimini, Bohigues, 2020: 634).

These authors identify trade liberalization, the degree of dependence of the internal market on manufactured products, and market concentration (i.e. the Herfindahl-Hirschman Index) as economic predictors of a preference for regionalism on the part of parliamentarians. Their analysis indicates that these factors affect policymakers’ preferences concerning economic regionalism in Latin America.

They found a negative correlation between trade opening and market concentration and support for economic regionalism among parliamentarians. By contrast, dependence on manufactured products correlates positively with support for regionalism. They conclude that economic constraints and incentives are as important for the formation of legislators’ preferences in terms of RFP as their political-ideological position.

Although plausible, we cannot infer that the economy matters in terms of the degree of involvement of Parliaments in Foreign Policy decisions. Furthermore, the study concerns Latin American countries and so does not allow for generalizing. Despite these limitations, the mentioned studies do offer an important insight into the relevance of economic context to the Legislature’s role in the formulation of Foreign Policy.

As we have seen, it is expected that institutional design and political conditions are relevant for explaining the participation of the Legislature, but the economy also plays a role. This base argument being closed, we can move on to the explanatory model.

Explanatory model

The question considered here is whether the participation of the Legislature in Regional Foreign Policy decisions is affected by domestic contexts. In the literature, the theme of legislative participation in international matters is controversial. Some authors suggest that national Parliaments play a significant role in Foreign Policy decision-making at a systemic level while others suggest the opposite (Oliveira, 2003; Onuki, Oliveira, 2006).

To prevent this controversy from muddying the waters, we asked specialists to point out cases that occurred in Mercosur and the EU that could be used to create an intentional sample for an empirical intermediate-N analysis. The intended advantage of this procedure is the reduction of possible biases in our case selection.⁴

Selected Cases

From the consultation, five decisions were selected for analysis: The Treaty of Asunción (1991), the Protocol of Ushuaia (1998), the Accession of Venezuela to Mercosur (2006), the Treaty of Maastricht (1992), and the Treaty of Lisbon (2007).

Considering this decision, the countries for analysis were selected using three criteria: (i) scenarios in which it is possible to examine whether the Foreign Policies of the States that makeup Mercosur and the EU, respectively, are affected and/or subject to the control of their Legislative Branches; (ii) cases that cover different institutional designs and political configurations; and (iii) occurrences with data availability.

Cases and outcomes⁵

As the purpose of this article is to verify which factors, or combination of factors, lead to a certain outcome, the participation of the Legislature in Regional Foreign Policy decisions, the use of configurational methods, in this case, QCA and CNA, is especially useful (Rihoux, Ragin 2009; Ambuehl, Baumgartner, 2019). Such an approach requires that the variables be transformed, converting them into values that can be processed by the method. This transformation can be into a crisp set (converting them to 0 or 1) or a fuzzy set (converting them into values from 0 to 1). Table 1 brings together all the cases included in the analysis and the aforementioned transformation of the categorical variable Legislative Participation into a crisp set.

Table 1
: Cases selected for analysis and results of interest.

Case number	Decision	Case	Legislative Participation	particip
1	Treaty of Asunción	Uruguay (1991)	Very participative	1
2		Brazil (1991)	Participative	0
3		Argentina (1991)	Ambiguous coding	-
4	Venezuela’s accession to Mercosur	Brazil (2006)	Very participative	1
5		Uruguay (2006)	Participative	0
6		Argentina (2006)	Participative	0
7		Paraguay (2006)	Very participative	1
8	Maastricht Treaty	Germany (1992)	Very participative	1
9		France (1992)	Very participative	1
10		Portugal (1992)	Participative	0
11		Holland (1992)	Participative	0
12		Denmark (1992)	Very participative	1
13		Spain (1992)	Participative	0
14		Luxembourg (1992)	Participative	0
15	Lisbon Treaty	Luxembourg (2007)	Participative	0
16		Germany (2007)	Very participative	1
17		Denmark (2007)	Participative	0
18		Holland (2007)	Very participative	1
19		France (2007)	Very participative	1
20		Spain (2007)	Very participative	1
21		Portugal (2007)	Participative	0
22	Ushuaia Protocol	Brazil (1998)	Participative	0
23		Argentina (1998)	Participative	0
24		Uruguay (1998)	Participative	0
25		Paraguay (1998)	No data	-

Source: author’s elaboration.

In total, there are five Foreign Policy decisions across different periods, involving 11 countries – four from Mercosur and seven from the EU. The selection of the European countries promoted variations in the systems to guarantee different configurations of the cases. Different standards of States were chosen States of great political and economic importance to the EU (e.g., France and Germany), as well as mid (e.g., Spain and Portugal) and small ones (e.g., Luxembourg). Besides that, the choice of these States also considered the availability of data.⁶ As our unit of analysis is determined by the involvement of a country in each decision (country/decision), some countries appear more than once, thus being cataloged as different cases. There are a total of 25 observations. Another relevant aspect of the selection of cases was the variation in the result of interest.

This outcome was operationalized based on the authors’ case studies and it was coded by category (participatory vs. very participatory). This means that, to some degree, the Legislature always participates in Foreign Policy decisions but there is a variation in the extent of said participation, which can be more or less intense depending on the case. For example, in the process of Venezuela’s accession to Mercosur, Brazil was cataloged as very participative because the discussions within the Legislature were intense, conflicting, and broad.

In this same process, the Argentine Legislature was defined as participative, as there was a greater degree of consensus among parliamentarians. Once presented by the Executive, the bill for approval, the adhesion protocol was ratified by the Chamber and the Senate in one month, leading to the promulgation of Law No. 26,192 of 2006.

In sum, the cases were cataloged as very participative when the path to the approval/refusal of a document extrapolated the supervising role of the Legislative branch and when the legislators faced low levels of consensus in the Houses. This is because the governments have incentives to establish legislative committees and procedures to avoid conflicts in foreign policy matters (Beijing Session, 1996). It means that we should expect regular participation of the Legislative branch in such a sector. However, even in cases of high institutionalization, as in Denmark (Beijing Session, 1996), there is variation in the degree of participation, showing that the consensus, by itself, does not explain all the parliamentary responses. In other words, we are observing the intensity of the activities developed by the Parliaments and not whether the legislators are veto players.

Thus, the operationalization of Legislative participation in Regional Foreign Policy decisions can be done in numerical terms, attributing the value 0 to the cases in which the Legislature participated in a discrete way, or less intensely, and 1 to the cases in which the Legislature participated intensely in the process.⁷ This variable, converted into numerical terms, was named particip.

It was not possible to code the participation of the Legislature in two cases, Paraguay in 1998 and Argentina in 1991. In the first case, the information is not available online. In the second, the data exists but it was not possible to achieve a secure codification as the available records were scarce and only indicated the legal path necessary for the approval of the decision, with no further details on how active the Legislative Power was in the process. This reduces our sample to 23 cases.

Data, variables, and calibration of causal conditions

To carry out the QCA, the selected data to operationalize the causal conditions were clustered into two blocs of variables: (i) institutional and political variables – which includes data on the system of government, number of legislative chambers, effective number of parties, and whether a prior debate was held in the Legislature; and (ii) economic variables - economic growth (GDP), market concentration (HHI), and commercial opening. Table 2 shows the variables of the institutional political block as well as their respective calibrations for causal conditions.

Table 2
: Block of institutional and political variables/causal conditions (calibrated)

Cases	System of government	presid	Uni- or bicameral	bicam	Prior debate	prior_deb	ENP	frag
Germany 1992	Parliamentary	0	Bicameral	1	Yes	1	3.2	0.08
Germany 2007	Parliamentary	0	Bicameral	1	Yes	1	4.1	0.44
Argentina 1991	Presidential	1	Bicameral	1	-	-	2.9	0.00
Argentina 1998	Presidential	1	Bicameral	1	No	0	3.3	0.10
Argentina 2006	Presidential	1	Bicameral	1	Yes	1	5.8	0.93
Brazil 1991	Presidential	1	Bicameral	1	No	0	8.7	1.00
Brazil 1998	Presidential	1	Bicameral	1	No	0	7.1	1.00
Brazil 2006	Presidential	1	Bicameral	1	Yes	1	9.3	1.00
Denmark 1992	Parliamentary	0	Unicameral	0	No	0	4.4	0.58
Denmark 2007	Parliamentary	0	Unicameral	0	Yes	1	5.3	0.86
Spain 1992	Parliamentary	0	Bicameral	1	Yes	1	2.9	0.00
Spain 2007	Parliamentary	0	Bicameral	1	Yes	1	2.5	0.00
France 1992	Semi-presidential	0	Bicameral	1	Yes	1	3.1	0.06
France 2007	Semi-presidential	0	Bicameral	1	Yes	1	2.5	0.00
Netherlands 1992	Parliamentary	0	Bicameral	1	No	0	3.8	0.27
Netherlands 2007	Parliamentary	0	Bicameral	1	Yes	1	5.5	0.89
Luxembourg 1992	Parliamentary	0	Unicameral	0	Yes	1	3.8	0.27
Luxembourg 2007	Parliamentary	0	Unicameral	0	No	0	3.8	0.27
Paraguay 1998	Presidential	1	Bicameral	1	-	-	2.0	0.00
Paraguay 2006	Presidential	1	Bicameral	1	No	0	3.2	0.08
Portugal 1992	Semi-presidential	0	Unicameral	0	No	0	2.2	0.00
Portugal 2007	Semi-presidential	0	Unicameral	0	No	0	2.6	0.00
Uruguay 1991	Presidential	1	Bicameral	1	Yes	1	3.3	0.10
Uruguay 1998	Presidential	1	Bicameral	1	No	0	3.3	0.10
Uruguay 2006	Presidential	1	Bicameral	1	No	0	2.4	0.00

Sources: World Bank and original data coded by the authors.

Institutional and political variables/causal conditions

Regarding institutional design, the first variable is the system of government, which we collected from the World Bank.⁸ It is divided into presidential, semi-presidential, and parliamentary. For calibration purposes, presidential systems were coded as 1, and semi-presidential and parliamentary systems as 0. This causal condition was denominated presid. For the variable uni- or bicameral, there was no need for calibration as it is naturally binary. Thus, bicameral countries were coded as 1 and unicameral countries as 0. This causal condition was denominated bicam.

The condition prior debate (prior_debat) was derived from case studies we conducted from documentation and records regarding the events cited in section 3.1. The operationalization was based on the presence/absence of references to the aforementioned decisions in the discussions documented before the year indicated as the initial point for each case (indicated in Table 2). We assigned a value of 1 for cases when there was debate in the Legislature before the initial year of the decision and 0 for cases where this did not happen. For example, even before 1991, there were discussions in Uruguay referring to the Treaty of Asunción, thus this case was cataloged as 1.

We calibrated all these variables as binary (or dichotomous) conditions. Thus, for these causal conditions, the value 1 designates membership in the set while 0 represents exclusion. For example, the causal condition presid 1 is for countries that have presidential systems while 0 indicates those that are not subject to these systems.

These variables are categorical and not ordinal. In other words, they cannot be numerically transformed according to a hierarchy of categories, either in order of magnitude or importance.

Finally, the effective number of parties⁹ is the basis of the causal condition “party fragmentation” (frag). In this case, the calibration took place in the form of a fuzzy set and met the following criteria: cases with ENP greater than 6 were considered members of the set with a totally fragmented Legislature and received a value of 1 (e.g., Brazil 1991, 1998, and 2006). We recorded cases with ENP less than 3 as not belonging to the set with fragmented Legislatures, receiving a value of 0 (e.g. France 2007 and Spain 2007).

The cases with ENP between 3 and 6 are considered according to their degree of membership in the set of fragmented Legislatures, giving the ENP of 4.2 (average ENP) as a cutoff point. So, cases with ENP between 3 and 4.2 more not belong than belong to the set of cases with fragmented Legislature (e.g., Netherlands 1992 and Germany 2007) than otherwise. Cases with values of ENP between 4.2 and 6 are more in than out of the set of cases with fragmented Legislatures (e.g., Netherlands 2007 and Denmark 1992).

Economic variables/causal conditions

We selected three economic variables: economic growth, trade openness, and market concentration (HH indicator). All were calibrated as fuzzy sets as they are better expressed in differences of degree and not in kind. Economic growth was operationalized as the average growth of each case over the last 4 years in percentage terms of GDP.¹⁰

This average growth was then calibrated so that cases with a total absence of growth were considered. Those with averages equal to or less than 0% received the value 0, which means that they are non-members of the set of economically growing countries. Cases with averages equal to or above 4% are coded as 1, which means that they are full members of the set of countries with growing economies. For cases with values between 0% and 4%, we established a threshold of 2%, thus above 0% and below 2% there are more non-members than members; above 2% and below 4%, there are more members than non-members. Table 3 shows the original and calibrated values, both for this variable and for the others.

Table 3
: Block of economic variables/causal conditions (calibrated).

Cases	Growth (average % GDP 4 years)	growth	Trade openness	trade_open	Market concentration (HH)	Mark_conc
Germany 1992	4.04	0.95	38.75	0.00	0.07	0.04
Germany 2007	2.17	0.56	77.72	0.15	0.04	0.00
Argentina 1991	-	-	17.69	-	-	-
Argentina 1998	3.66	0.92	30.55	0.00	0.12	0.42
Argentina 2006	8.69	1.00	32.35	0.00	0.06	0.00
Brazil 1991	0.39	0.09	10.17	0.00	0.14	0.54
Brazil 1998	2.58	0.70	16.32	0.00	0.07	0.04
Brazil 2006	3.51	0.90	20.28	0.00	0.06	0.00
Denmark 1992	1.36	0.28	56.96	0.00	0.18	0.71
Denmark 2007	2.45	0.66	95.44	0.33	0.07	0.04
Spain 1992	3.02	0.82	30.79	0.00	0.16	0.63
Spain 2007	3.62	0.92	58.19	0.00	0.07	0.04
France 1992	2.47	0.67	33.37	0.00	-	-
France 2007	2.34	0.62	55.22	0.00	0.06	0.00
Netherlands 1992	3.18	0.85	79.28	0.16	0.31	1.00
Netherlands 2007	2.81	0.77	130.32	0.57	0.10	0.27
Luxembourg 1992	6.39	1.00	171.23	0.74	-	-
Luxembourg 2007	5.07	1.00	300.58	1.00	0.09	0.19
Paraguay 1998	-	-	70.76	-	-	-
Paraguay 2006	3.82	0.94	73.46	0.10	0.10	0.27
Portugal 1992	3.96	0.95	42.70	0.00	0.19	0.75
Portugal 2007	1.67	0.38	67.78	0.05	0.11	0.35
Uruguay 1991	1.60	0.36	27.06	0.00	-	-
Uruguay 1998	4.29	1.00	37.77	0.00	0.16	0.63
Uruguay 2006	4.34	1.00	50.35	0.00	0.05	0.00

Source: World Bank.11

Trade openness is calculated as follows: (exports + imports)/GDP. This measures the country’s exposure to international trade, indicating the degree of dependence of domestic producers on the foreign market as well as of domestic consumers on foreign products. The greater the value of trade openness, the more significant international trade is for the economy, with records above 100 indicating that foreign trade exceeded the country’s GDP.

As for the market concentration, we opted for the Herfindahl-Hirschman Indicator which measures the dispersion of the country’s export volume among trading partners. The more concentrated the export destination (few partners), the greater the country’s dependence on a limited number of partners. In this case, the indicator value will be closer to 1. The greater the diversity of export destinations, the closer the concentration indicator will be to 0.

For calibration of the conditions of commercial openness and market concentration, the findTH function was used from the QCA package for R (Dușa, 2019) in standard definition, which uses the technique of complete hierarchical groupings according to Euclidean distance. In this definition, cluster analyses are used to establish which calibration anchors best separate the observed cases depending on the chosen condition. Such a procedure is recommended only in the absence of a theoretical framework on how a condition should be calibrated (Dușa 2019); that is when there are no theoretically oriented expectations regarding qualitative transformations in the cases analyzed according to their levels of belonging to the condition in question.

Qualitative Comparative Analysis (QCA) and Coincidence Analysis (CNA)

To answer the research question “What conditions determine the level of participation of national Legislatures?” this study employs QCA to empirically examine which configurations out of institutional design, political conditions, and economic factors lead to increased participation of Legislatures in Regional Foreign Policy.

The model

We therefore expect that factors related to the institutional design, and the political and economic context of each case will be decisive for the participation of the Legislature in Foreign Policy decisions. In more appropriate methodological terms for QCA, we expect to find the necessary conditions as well as sufficient causal configurations to explain the participation of the Legislature in Foreign Policy decisions. The operationalization of the model can be represented as follows.

$\text{Particip}=f{\left[\text{presid}{}^{\star }{\text{bicam}}^{\star }\text{frag*}{}^{\text{prior\_deb*}}\text{(growth+trade\_open+mark\_conc)}\right]}^{}$ ¹²

Where particip is the result of interest, which must be a function of the terms on the right side of the equation. The first two are characteristics of institutional design presid and bicam, while the terms frag and prior_deb represent political aspects.

Note that the three economic conditions appear in parentheses. They are growth, trade_open, and mark_conc. This is due to the strategy adopted in the analysis, which aims to test each of them separately but in conjunction with institutional and political conditions. Thus, we have four models: Model 1 (with institutional and political variables only); Model 2 (with institutional and political variables, and economic growth); Model 3 (with institutional variables and trade liberalization policies); and, finally, Model 4 (with institutional variables and market concentration).

Table 4 shows the analytical matrix, duly calibrated and suitable for analysis. It includes the result of interest (Y), the institutional and political causal conditions (X1, X2, X3, and X4), and the economic causal conditions (X5a, X5b, and X5c).

Table 4
: Analytical matrix

	Institutional and political conditions					Economic conditions
Case	particip (Y)	presid X1	bicam X2	frag X3	prior_deb X4	growth X5a	trade_open X5b	Mark_conc X5c
Germany 1992	1	1	1	0.08	1	0.95	0.00	0.04
Germany 2007	1	0	1	0.44	1	0.56	0.15	0.00
Argentina 1998	0	1	1	0.10	0	0.92	0.00	0.42
Argentina 2006	0	1	1	0.93	1	1.00	0.00	0.00
Brazil 1991	0	1	1	1.00	0	0.09	0.00	0.54
Brazil 1998	0	1	1	1.00	0	0.70	0.00	0.04
Brazil 2006	1	1	1	1.00	1	0.90	0.00	0.00
Denmark 1992	1	0	0	0.58	0	0.28	0.00	0.71
Denmark 2007	0	0	0	0.86	1	0.66	0.33	0.04
Spain 1992	0	0	1	0.00	1	0.82	0.00	0.63
Spain 2007	1	0	1	0.00	1	0.92	0.00	0.04
France 1992	1	0	1	0.06	1	0.67	0.00	-
France 2007	1	0	1	0.00	1	0.62	0.00	0.00
Netherlands 1992	0	0	1	0.27	0	0.85	0.16	1.00
Netherlands 2007	1	0	1	0.89	1	0.77	0.57	0.27
Luxembourg 1992	0	0	0	0.27	1	1.00	0.74	-
Luxembourg 2007	0	0	0	0.27	0	0.99	1.00	0.19
Paraguay 2006	1	1	1	0.08	0	0.94	0.10	0.27
Portugal 1992	0	0	0	0.00	0	0.95	0.00	0.75
Portugal 2007	0	0	0	0.00	0	0.38	0.05	0.35
Uruguay 1991	1	1	1	0.10	1	0.36	0.00	-
Uruguay 1998	0	1	1	0.10	0	0.97	0.00	0.63
Uruguay 2006	0	1	1	0.00	0	0.97	0.00	0.00

Source: author’s elaboration.

It is important to note that, depending on the conditions introduced in the different models, there can be losses in some cases. We are working with 23 cases, which drops to 20 cases in the model that considers economic concentration.

The QCA approach is prodigal in identifying necessary conditions and sufficient causal configurations, but it is not the best way to discern causal chains since it uses the Quine-McCluskey algorithm for logical minimization. Furthermore, it relies on counterfactual conjectures that are often unsustainable for parsimonious solutions (Marisguia, 2020). Thus, we not only make use of QCA but also CNA.

Results

For the QCA, we will follow the more conventional protocol which suggests that we start with the analysis of necessary conditions and then move on to that of sufficient conditions. We will consequently proceed with the CNA.

Necessary Conditions (QCA)

The analysis allowed for the identification of some necessary conditions, with variations between the models. Table 5 presents a summary of these findings. Bicameralism appears as a necessary condition in three of the four models. In Model 4, however, it appears ambiguously, first as a negative and second because it only appears to be necessary if combined with other conditions.

Table 5
: Analysis of Necessary Conditions

Model 1 – Only with institutional and political conditions.
Conditions	Consistency	Relevance
bicam	0.9	0.42
~pres + prior_deb	0.9	0.42
~bicam + prior_deb	0.9	0.50
Model 2 - Institutional and political conditions + economic growth.
Conditions	Consistency	Relevance
bicam	0.9	0.42
~pres + prior_deb	0.9	0.42
~bicam + prior_deb	0.9	0.50
Model 3 - Institutional and political conditions + trade openness.
Conditions	Consistency	Relevance
bicam	0.9	0.42
~trade_open	0.9	0.22
~pres + prior_deb	0.9	0.42
~bicam + prior_deb	0.9	0.50
Model 4 - Institutional and political conditions + market concentration.
Conditions	Consistency	Relevance
~pres + prior_deb+ mark_conc	0.9	0.32
~bicam + prior_deb+ mark_conc	0.9	0.32
~bicam + ~mark_conc	0.9	0.30

Source: author’s elaboration.

Note that the consistency statistic is 0.9 and, according to the literature, in models with conditions like fuzzy any value from 0.9 can be considered an almost necessary condition.¹³ It is important to note that bicameralism can be considered, strictly speaking, an almost necessary condition as it has a consistency of 0.9, but its importance must be qualified in the explanation. The relevance of this necessary condition is only 0.42. Although there is no determined limit value in the literature to consider a necessary condition as effectively relevant, as it is a statistic that varies from 0 to 1 and is of more relevance the closer it is to 1, the value can be considered low.¹⁴ This is because most countries in the sample are bicameral, with the exceptions of Portugal, Denmark, and Luxembourg, as shown in the data matrix (Table 4).

Finally, it is worth noting that Model 4, which includes the economic condition market concentration, differs substantially from the others. There is no clear explanation for this result but it is quite likely that it is related to the loss of some cases – three cases to be precise – due to a lack of data. In any case, this model does not permit us to identify any necessary conditions in isolation.

Sufficient Conditions (QCA)

Table 6 shows the results of the sufficiency analysis. In Model 1, the complex solution¹⁵ shows two sufficient settings. According to this model, without an economic variable, an increased participation of the Legislature occurs: (i) in non-presidential democracies associated with the presence of bicameralism and prior debates; or (ii) in bicameral countries, as long as there is prior debate and no party fragmentation. Coverage is satisfactory at 0.69, Model 1 being the model that presented the best coverage.

Table 6
: Analysis of QCA sufficient conditions

Model 1 – With institutional and political conditions only
Complex solution	Cons.	Unique Cov.	Cov	7 cases
~pres*bicam*prior_deb	0.85	0.10	0.60	Germany_1992, Germany_2007, Spain_1992, Spain_2007, France_1992, France_2007
bicam*prior_deb*~frag	0.84	0.00	0.54	Germany_1992, Germany_2007, Spain_1992, Spain_2007, France_1992, France_2007
Total solution	0.86	-	0.69	Germany_1992, Germany_2007, Spain_1992, Spain_2007, France_1992, France_2007 and Netherlands_2007
Model 2 - Institutional and political conditions + economic growth
Complex solution	Cons.	Unique Cov.	Cov.	9 cases
~pres*bicam*prior_deb*crescim	0.85	0.45	0.45	Germany_1992, Germany_2007, Spain_1992, Spain_2007, France_1992, France_2007
pres*bicam*prior_deb*~frag*~crescim	1.00	0.10	0.10	Uruguay_1991
~pres*~bicam*~prior_deb*frag*~crescim	0.94	0.10	0.1	Denmark_1992
Total solution	0.87	-	0.57	Germany_1992, Germany_2007, Spain_1992, Spain_2007, France_1992, France_2007, Netherlands_2007, Uruguay_1991 and Denmark_1992
Model 3 - Institutional and political conditions + trade openness
Complex solution	Cons.	Unique Cov.	Cov.	9 cases
bicam*prior_deb*~frag*~abert_com	0.84	0.51	0.54	Germany_1992, Germany_2007, Spain_1992, Spain_2007, France_1992, France_2007 and Uruguay_2007
~pres*bicam*prior_deb*frag*abert_com	1.00	0.04	0.10	Netherlands_2007
~pres*~bicam*~prior_deb*frag*~abert_com	0.96	0.10	0.10	Denmark_1992
Total solution	0.86	-	0.64	Germany_1992, Germany_2007, Spain_1992, Spain_2007, France_1992, France_2007, Netherlands_2007, Uruguay_1991 and Denmark_1992
Complex solution	Cons.	Unique Cov.	Cov.	5 cases
~pres*bicam*prior_deb*~mark_conc	0.92	0.92	0.58	Germany_1992, Germany_2007, Spain_2007, France_2007 and Netherlands_2007

Source: author’s elaboration.

Model 2, with the variable economic growth, provides for a more complex interpretation with three sufficient configurations: (i) non-presidentialism associated with bicameralism, prior debate, and economic growth; (ii) bicameral presidentialism associated with prior debates, no fragmented legislature, and no growth; and (iii) unicameral non-presidentialism (the case of Denmark), without prior debate, fragmented legislature, and with no growth. The coverage of Model 2 is the lowest among all the models considered. This shows that the introduction of the variable economic growth is of no use in providing an elegant explanation for the phenomenon.

Model 3 is the least parsimonious, showing three quite extensive configurations for the explanation. It is the one with the best coverage (0.64) in the complex solution but, like the previous model, it is complicated to interpret. It is important to note that two of the three configurations are sufficient to explain only one case each and do not add much information to the more general analysis. As in Model 2, the inclusion of the variable commercial opening does not contribute decisively to the analysis of the phenomenon.

Finally, Model 4 shows a very interesting complex solution compared to the others. It reveals just one sufficient configuration. For this solution, the Legislature participates when the country is non-presidential, bicameral, has prior debate in the Legislature, and has no market concentration. With only one causal configuration, the model achieves a coverage of 0.58, considered quite reasonable. Yet the coverage is inferior if compared to the previous model.

The comparison of the models proved to be quite enlightening: at least three relevant findings can be listed.

The first is that the participation of the Legislative in Foreign Policy decisions is a characteristic of non-presidential countries, which are, in their great majority, European countries. The verification of the cases shows that the only exception is Uruguay in 1991 which is covered by Models 2 and 3 - not a surprising find as Uruguay is an outlier case in Latin America. The result is compatible, on the one hand, with the tradition of European countries in Regional Foreign Policy, which have had different forms of regionalism since the 1950s when the European Coal and Steel Community was established. On the other hand, concerning Uruguay, the finding also adjusts to the fact that this country has the most consensual form of presidentialism in Mercosur. In other words, the power is not concentrated in the Executive but broadly divided among the Branches, and this can catalyze legislative participation (Oliveira, 2006).

This finding raises a problem for our aspiration to present a generalizable explanation of legislative participation in RFP decisions. As we have seen, the models explain European countries well but not Latin American ones. It is worth emphasizing that among the nine Latin American cases there was intense legislative participation in three. And the explanation for this participation remained pending. Certainly, more theoretical and empirical efforts need to be made to include other variables with explanatory capacity for the greatest number of cases in the model. This suggests that, in Latin America, other causal conditions not included in the model may be decisive. Perhaps the tradition of presidential diplomacy helps to better understand the context in the region - but this is just speculation. Future studies need to be carried out to explore other possibilities.

The second finding is that the inclusion of economic conditions proved to be valuable for the analysis. They revealed a greater diversity and allowed us to explain a higher contingent of cases, especially Model 4 which includes the condition market concentration. This model, as we have seen, presented a very elegant solution with coverage that can be considered reasonable. It is true that this model was somewhat impaired and differed from the others, probably due to the loss of three cases for which no data were found. Still, its results are relevant.

Third, but not least, is that bicameralism and prior debate seem to be very relevant conditions to explain the intense participation of the Legislative in Foreign Policy decisions. These two conditions can be identified in seven of the nine sufficient causal configurations found in the four models.

Coincidence Analysis (CNA)

To sequentially connect causal conditions and thereby determine a chain that favors the understanding of conditions that precede others, we employed the comparative configurational method called CNA.

This method aims to identify INUS conditions,¹⁶ but it differs from QCA in that it does not use the Quine-McCluskey algorithm. The search for causal chains is done with the artifice of considering causal dependencies not just between an outcome Y and Xn conditions, but also among all the factors included in the model. This is to identify causal sequences between the variables that led to the result of interest, a function absent in QCA (Baumgartner, 2015 apud Marisguia, 2020).

For example, we know that institutional factors like presidentialism, bicameralism, and party fragmentation have a certain perennial quality. However, political factors are cyclical. So, a possible question would be: do bicameral non-presidential systems provoke greater involvement of the Legislature in a wider range of discussions, provoking prior debate that leads the Legislature to participate intensely in Foreign Policy decisions?

As CNA can test all conditions as outcomes indiscriminately, it tends to identify many causal chains that, despite their good consistency and coverage, do not make theoretical sense. For example, treating bicameralism as an outcome, one can conclude that growth*~pres cause bicameralism, which, of course, does not make any sense at all. Seeking to reduce the complexity of the analysis and maintain consistency, we assume that there are only two possible outcomes that make sense:¹⁷ legislative participation and prior debate. We start, therefore, from the premise that the other conditions can only be causal and never the result of interest. Furthermore, in seeking to reduce the complexity of the analysis we have only reported the causal chains that make theoretical sense. Another criterion adopted to guide our study and the way we report our results was to establish a threshold of 0.75 for consistency and coverage, as recommended by Ambuehl & Baumgartner (2019).

The CNA undertaken here presented relevant results. Table 7 summarizes the results for Models 1, 2, and 3 due to the similarity of the solutions between them. Table 8 shows the results of Model 4 which, due to its peculiarity and greater relevance, is analyzed separately.

Table 7
: Analysis of CNA causal chains

Model 1 – With institutional and political conditions only
Outcome	Atomic solutions	Consistency	Coverage	Cases
particip	bicame*deb.prev	0.80	0.80	Germany_1992, Germany_2007 Argentina_2006, Brazil_2006, Spain_1992, Spain_2007, France_2007, France_1992, Netherlands_2007, Uruguay_1991
Model 2 - Institutional and political conditions + economic growth
particip	bicame*deb.prev	0.80	0.80	Germany_1992, Germany_2007 Argentina_2006, Brazil_2006, Spain_1992, Spain_2007, France_2007, France_1992, Netherlands_2007, Uruguay_1991
Model 3 - Institutional and political conditions + trade openness
particip	bicame*deb.prev	0.80	0.80	Germany_1992, Germany_2007 Argentina_2006, Brazil_2006, Spain_1992, Spain_2007, France_2007, France_1992, Netherlands_2007, Uruguay_1991

Source: author’s elaboration.

Table 8
: Analysis of CNA causal chains

Model 4 - Institutional and political conditions + market concentration.
Outcome	Atomic solutions	Consistency	Coverage	Cases
prior_deb	~pres*bicam + ~pres*frag*~mark_conc	0.81	0.76	Germany_1992, Germany_2007, Denmark_2007, Spain_1992, Spain_2007, France_2007, Netherlands_1992, Netherlands_2007
particip	prior_deb*bicam	0.75	0.75	Germany_1992, Germany_2007, Brazil_2006, Argentina_2006, Spain_1992, Spain_2007, France_2007, Netherlands_2007
Complex solution
~pres*bicame + ~pres*frag*~mark_conc <-> prior_deb * prior_deb*bicame <-> par_legis				Germany_1992, Germany_2007, Argentina_2006, Brazil_2006, Denmark_2007, Spain_1992, Spain_2007, France_2007, Netherlands_1992, Netherlands_2007

Source: author’s elaboration.

Briefly looking at the solutions of Models 1, 2, and 3 we can see that they did not identify causal chains and provided only atomic solutions. Such solutions suggest that bicameralism associated with prior debate in the Legislature are causal conditions that lead to legislative participation, no matter what the governmental system is (i.e., presidentialism or non-presidentialism), regardless of the degree of fragmentation of the Legislature. These solutions also show us that the conditions of economic growth and commercial opening are not relevant for the participation of the Legislature. None of them appears as a prime implicant.

Considering this result, we note that the variables economic growth and commercial opening cannot be considered prime implicants. Even so, the result of Model 4, which includes the condition market concentration, draws attention to the relevance of introducing economic variables in the interpretation of Legislative participation in RFP decisions.

It should be noted that in this model the CNA identified two causal chains. The first suggests that non-presidential bicameral countries or non-presidential countries with less fragmented legislatures and little market concentration lead to prior debate. The second suggests that prior debate with bicameralism leads to the participation of the Legislature.

This solution covers 10 cases, including countries like Argentina and Brazil in 2006. Until then, the QCA presented only Uruguay in 1991 as a case covered by the various solutions presented. If we look at Table 4 we see that, of the 10 cases covered by this model’s complex solution, only three did not have intense participation by the Legislature in the decision. They were Argentina 2006, Denmark 2007, and Spain 1992. This, therefore, seems to us the most enlightening solution.

Summary of findings

The analysis of necessary conditions identified bicameralism as a condition without which intense participation on the part of the Legislature is unlikely to occur. However, bicameral legislatures constitute the vast majority in our sample, which weakens their explanatory relevance. Only Portugal, Denmark, and Luxembourg are not bicameral. Careful inspection of Table 4 reveals another condition that seems relevant: prior debate. Note that this condition appears in several necessary configurations. Thus, although it does not have the status of a necessary condition, it appears to be a rather important functional equivalent.

When comparing the analysis of necessity with that of sufficiency, a compatible result is found. Bicameralism and prior debate are both necessary (or almost necessary) and very relevant sufficient conditions to explain the intense participation of the Legislature in RFP decisions. These two conditions can be identified in seven of the nine sufficient causal configurations found in the study by QCA.

The CNA significantly added to the analytical strategy undertaken in at least three aspects: it confirmed the importance of bicameralism and prior debate for the explanation, it showed the importance of including the economic variable market concentration, and it helped to identify at least one relevant causal chain.

Regarding the importance of the economic variable, specifically market concentration, we find that countries with a lack of market concentration (in combination with other conditions) tend to have more participatory legislatures. The less concentrated the number of export destinations, the greater the number of trading partners the country has. This leads to less dependence on a limited number of countries. This history of commercial diversification is quite probably related to the performance of the Parliament.

Finally, concerning the identification of the causal chain, an important finding was that prior debate leads to more intense participation by the Legislature. Substantially, this allows us to state that the Legislature having time and a tradition of engagement in Foreign Policy debates, as is the case in European countries, is essential for it to participate intensively in Foreign Policy decision-making. This finding seems obvious but it is not. The Legislature not acting as a protagonist and/or not being systematically involved in the field weakens it when it participates in Foreign Policy decisions.

Not without reason, the model does not explain attitudes to RFP in Latin American countries, where presidential diplomacy is a prominent fact and regionalism lacks the tradition and depth it enjoys in the European model. Uruguay differs from other Mercosur members in that it has a more consensual and compromise-driven presidential system (Oliveira, 2006). The Uruguayan presidency is underwritten by Executive-Legislative agreements without the establishment of coalitions (Lanzaro, 2012). These aspects cause Uruguayan Foreign Policy to deviate from the norm of Latin American presidential diplomacy, as it is more centered on a model of rule by party, in which political parties play a significant role in decision-making thus boosting the role of the Legislature (Burian, 2014; Reimberge, Andrade, 2020).

The results reported in Table 7 below are those characterized by better consistency and coverage, which make the most sense theoretically. For a complete overview of all results, the data and scripts will be available for replication.

Table 9
: Analysis of CNA sufficient conditions

Model 1 – With institutional and political conditions only
Solution		Consistency	Coverage	Cases
bicam*prior_deb		0.8	0.8	Germany_1992, Germany_2007 Argentina_2006, Brazil_2006, Spain_1992, Spain_2007, France_2007, France_1992, Netherlands_2007, Uruguay_1991
Model 2 - Institutional and political conditions + economic growth
Solution		Consistency	Coverage	Cases
bicam*prior_deb		0.8	0.8	Germany_1992, Germany_2007 Argentina_2006, Brazil_2006, Spain_1992, Spain_2007, France_2007, France_1992, Netherlands_2007, Uruguay_1991
Model 3 - Institutional and political conditions + trade openness
Solution		Consistency	Coverage	Cases
bicam*prior_deb		0.8	0.8	Germany_1992, Germany_2007 Argentina_2006, Brazil_2006, Spain_1992, Spain_2007, France_2007, France_1992, Netherlands_2007, Uruguay_1991
Model 4 - Institutional and political conditions + market concentration.
Outcome	Solution	Consistency	Coverage	Cases
prior_deb	~pres*bicame ~pres*frag*~mark_conc	0.81	0.76	Germany_1992, Germany_2007, Denmark_2007, Spain_1992 Spain_2007, France_2007 Netherlands_1992 and Netherlands_2007
par_legis	prior_deb*bicam	0.75	0.75	Germany_1992, Germany_2007, Brazil_2006, Argentina_2006, Spain_1992, Spain_2007, France_2007 and Netherlands_2007

Source: author’s elaboration.

Conclusions

To investigate which configurations of institutional design, political conditions, and economic factors lead to increased participation of the Legislature in Regional Foreign Policy, we proposed a QCA and CNA of the actions of States in five decisions related to Mercosur and the European Union: the Treaty of Asunción, the Adhesion of Venezuela to Mercosur, the Ushuaia Protocol, the Treaty of Maastricht the Treaty of Lisbon. Such decisions were selected based on a consultation with specialists.

The choice of countries considered in each decision mentioned above was due to the scope of different situations, the possibility of verifying whether the various foreign policies were subject to legislative control, and, finally, the availability of data.

The results showed that legislative moderation in matters of Regional Foreign Policy works as an instrument of legitimacy. In other words, the participation of national Legislatures in Mercosur and EU member countries is imperative and only varies in intensity. The Legislature can act directly, during the process of approving agreements, or indirectly, such as when changes in domestic legislation are requested or even when mechanisms for congressional monitoring are established (Pinheiro, 2008). Through these various and nuanced mechanisms, they can legitimate and exert influence over processes of international cooperation. Otherwise, when such instruments have their legitimacy denied, the credibility of the foreign policy and democratic commitments is undermined (Martin, 2000).

Furthermore, at the systemic level, governments seek to satisfy internal pressures and are affected by domestic information and preference structures (Putnam 1993; Milner 1997). This being the case, the question remains: What conditions determine the level of participation of national Legislatures? Domestic institutional design, which underlies the interaction between the Executive and Legislative Branches of Government, as well as the political relations between them, were the first conditions considered in this article.

This is because there is an expectation in the literature that the Legislature uses institutional mechanisms to control and monitor the actions of the Executive (Anastasia, 2012; Feliú, 2018). Nonetheless, such instruments are not the same for all States, since in each case the branches of government deal with different organizational systems and structures (Llanos, Nolte, 2003; Pinheiro, 2008; Lijphart, 2012). These configurations underlie different mechanisms of checks and balances which, in turn, affect the formation, level of fragmentation, and performance of coalitions and political parties (Onuki, Oliveira, 2010; Ribeiro, Pinheiro, 2016; Ribeiro, Oliveira, 2018).

These arguments were corroborated by three of the four CNA condition analysis models, which indicate that given institutional, political, and economic conditions, bicameralism constitutes a necessary condition (and constitutes a key sufficient configuration alongside prior debate in the legislature) to understand the intense participation of the Legislature in RFP decision-making.

Another condition considered for our analysis was the economic scenario, as studies have shown that the economy has a significant impact on the behavior of governments and the conduct of foreign policy (McAvoy, 2015). Hence, it accounts for a salient theme for the participation of actors other than the Executive. Trade openness, dependence on the internal market for manufactured products, and market concentration work as predictors of parliamentary preferences (Santos, Cimini, Bohigues, 2020).

Therefore, the inclusion of economic conditions allowed us to explain a higher number of cases. Model 4 (Table 6) corroborated our expectation about one of the economic variables: the market concentration index. Countries with a lack of market concentration tend to have a Legislature that participates more in RFP, but this condition must be associated with others such as non-presidentialism, bicameralism, and prior debate.

In short, the models demonstrated that institutional and economic matters affect legislative participation in Regional Foreign Policy. However, our results have greater explanatory power for European cases. This is justified by the fact that Europe has a long history of RFP, which makes the participation of domestic entities more concrete. Further studies are needed to understand the conditions and phenomena of Latin American Regional Foreign Policy.

Finally, this article demonstrates the relevance of the relationship between the analysis of institutions by Comparative Politics and International Politics.¹⁸ The synergy of these areas sheds light on political events, allowing for a better understanding of them. Hence, further analysis is important to fully understand the motivations behind the legislative engagement in foreign policy. We encourage research agendas encompassing additional dimensions of legislative participation in foreign policy, including parties, specialized committees, and individual legislators. Additionally, we aim to inspire researchers to analyze more deeply the Latin American Regional Foreign Policy and use interdisciplinary approaches to explain political phenomena.

Acknowledges

*We are thankful to Dawisson Belém Lopes (UFMG), Rafael Mesquita (UFPE), and Breno Marisguia (UFMG) for their comments and suggestions to the first drafts of this article. We appreciate the support of the Department of Political Science (Federal University of Minas Gerais - UFMG) with funding for the official translation of this paper. This research was funded by the Brazilian National Council for Scientific and Technological Development (CNPq), 303247/2015-0, and the Foundation for the Support of Science and Technology in the State of Pernambuco (FACEPE), IBPG-0312-7.09/20.

References

• Almond, Gabriel A. (1956), "Public opinion and national security policy". Public Opinion Quarterly, v. 20, n. 2, pp. 371-378
• Anastasia, Fátima; Mendonça, Christopher; Almeida, Helga. (2012), "Poder legislativo e política externa no Brasil: jogando com as regras". Contexto Internacional, v. 34, n. 2, pp. 617-657.
• Arendt, Hannah. (1998), "O que é política?", trad. R. Guarany, Rio de Janeiro, Bertrand Brasil.
• Ambühl, Mathias; Baumgartner, Michael. (2019), NA: Causal Modeling with Coincidence Analysis. R package version 2.2.1. Available on https://CRAN.R-project.org/package=cna
  » https://CRAN.R-project.org/package=cna
• Beijing SESSION. (1996). Role of the Parliament in the conduct of foreign relations. Report prepared by Mr. Seppe Tiitinen (Finland), adopted at the Beijing Session. Available on: https://www.asgp.co/sites/default/files/documents//OKHKAJPGHKJSCUHUKYOJUHFKTKJEVL.pdf
  » https://www.asgp.co/sites/default/files/documents//OKHKAJPGHKJSCUHUKYOJUHFKTKJEVL.pdf
• Betarelli Junior, Admir Antonio; Ferreira, Sandro de Freitas. (2018), Introdução à análise qualitativa comparativa e aos conjuntos Fuzzy (fsQCA), Brasília, ENAP.
• Caporaso, James A. (1997), "Across the Great Divide: Integrating Comparative and International Politics". International Studies Quarterly, v. 41, pp. 563-592.
• Carter, Ralph G.; Scott, James M. (2012), "Congress" in Steven W. Hook & Christopher M. Jones (Eds.) Routledge handbook of American foreign policy, New York, Routledge.
• Dahl, Robert A.; Shapiro, Ian.; Hacker-Cordon, Casiano. (1999), "Can international organizations be democratic? A skeptic's view". The Cosmopolitanism Reader, pp.19-36.
• De Mesquita, Bruce Bueno; Smith, Alastair. (2012), "Domestic explanations of international relations". Annual Review of Political Science, v. 15, pp. 161-181.
• Diniz, Simone; Ribeiro, Cláudio Oliveira. (2008), "The role of the Brazilian Congress in foreign policy: an empirical contribution to the debate". Brazilian Political Science Review (Online), v. 3, 0-0.
• Du?a, Adrian. (2019), "Critical tension: Sufficiency and parsimony in QCA". Sociological Methods & Research, v. 51, n. 2, pp. 541 - 566
• Feliú, Pedro. (2018), "Sobreviviendo a reformas, impopularidad y casos de corrupción: el presidencialismo de coalición del Brasil de Temer". Revista de ciencia política. Santiago, v. 38, n. 2, pp. 181-206, 2018
• Frizzera, Guilherme. (2013), "A suspensão do Paraguai no Mercosul: problema interno, solução externa". Conjuntura Global, v. 2, n. 3.
• FONSECA Jr, Gelson. (1998), A legitimidade e outras questões internacionais. São Paulo: Paz e Terra, pp. 293-352.
• Follesdal, Andreas., & Hix, S. (2008). "Candidate Selection" in T. Burean, H. Lupu, G. Badescu (eds.), The 2014 European Parliament Election In Poland. Romania, Babes-Bolyai University, pp. 129.
• Insaurralde, Gustavo. (2014), "Oposición parlamentaria y política exterior El rechazo del Partido Colorado al ingreso de Venezuela al MERCOSUR durante el gobierno de Fernando Lugo (2008-2012)". estudos internacionais v. 2, n. 2, pp. 149-187.
• Lantis, Jeffrey. S. (2006). "The life and death of international treaties: double-edged diplomacy and the politics of ratification in comparative perspective". International Politics, 43, pp. 24-52.
• Lanzaro, Jorge. (2010). "Uruguay: Presidencialismo pluralista y alternativas mayoritarias" in Lanzaro, Jorge, Presidencialismo y Parlamentarismo. América Latina y Europa Meridional, Centro de Estudios Políticos y Constitucionales, Madrid.
• Llanos, Mariana; Nolte, Detlef.; Weber, Cordula. T. (2012), "Paraguai: golpe ou voto de desconfiança?". Conjuntura Austral, v. 3, n. 14, pp. 4-17.
• Llanos, Mariana; & Nolte, Detlef. (2003), "Bicameralism in the Americas: around the extremes of symmetry and incongruence". The journal of legislative studies, v. 9, n. 3, pp. 54-86.
• Lima, Maria Regina S.; Santos, Fabiano. (2001), "O congresso e a política de comércio exterior". Lua Nova: Revista de Cultura e Política, v. 52, pp. 121-149.
• Lijphart, Arendt. (2012), Patterns of democracy: Government forms and performance in thirty-six countries, Yale University Press.
• Majone, Giandomenico. (1998), "Europe's 'democratic deficit': The question of standards". European law journal, v. 4, n. 1, pp. 5-28.
• Marisguia, Breno H. (2020), Predizendo e explicando interrupções presidenciais na América Latina. (Mestrado em Ciência Política), Universidade Federal de Minas Gerais.
• Martin, Lisa. (2000), "Democratic commitments" in L. Martin, Democratic Commitments, Princeton, Princeton University Press.
• Mcavoy, Gregory E. (2015), "Stability and Change: The Time Varying Impact of Economic and Foreign Policy Evaluations on Presidential Approval". Political Research Quarterly, v. 59, n. 1, pp. 71-83.
• Lessa, Antônio Carlos (2003). A Construção da Europa: a última utopia das Relações Internacionais. Brasília: Funag/IBRI.
• Mariano, Karine P.; Bressan, R.N.; y Luciano, B.T. (2017). "A comparative reassessment of regional parliaments in Latin America: Parlasur, Parlandino and Parlatino". Rev. Bras. Polít. Int, Brasília, v. 1, n. 60, pp. 1-18.
• Medeiros, Marcelo de Almeida.; Meunier, I.; y Cockles, Mariana. (2015). "Processos de Difusão Política e Legitimidade no Mercosul: Mimetismo Institucional e Mecanismos de Internalização de Normas Comunitárias". Contexto int., v. 37, n. 2, pp. 537-570.
• Medeiros, Marcelo de Almeida; Amelotti, Luiza Vilela; Moura, Nayanna Sabiá de (2021). "Representación ciudadana en el Parlasur (2011-2014): Un análisis comparativo entre el nivel doméstico y regional". Revista SAAP, v. 15, n. 2, pp. 478-509.
• Melo, Carlos Ranulfo. (2016), "Por que chegamos a tanto e que importância isso tem? Considerações sobre a fragmentação partidária no Brasil" in G. Perlin e M. L. Santos, Presidencialismo De Coalizão Em Movimento, 2019.
• Milner, Hellen V. (1997), Interests, institutions, and information: Domestic politics and international relations. Princeton University Press.
• Oliveira, Luzia Helena Herrmann de. (2006), "Presidencialismos em perspectiva comparada: Argentina, Brasil e Uruguai". Dados, v. 49, pp. 301-343.
• Oliveira, A. J.; Onuki, J. (2007). "Grupos de interesses e a política comercial brasileira: a atuação na arena legislativa" Papeis Legislativos, n. 8.
• Oliveira, Amâncio Jorge; Onuki, Janina. (2010), "Eleições, partidos políticos e política externa no Brasil". Revista Política Hoje, v. 19, n. 1
• Onuki, Janina; Ribeiro, Pedro Feliú; Oliveira, Amâncio Jorge. (2009), "Political parties, foreign policy and ideology: Argentina and Chile in comparative perspective". Brazilian Political Science Review (Online), v. 4.
• Onuki, Janina; Oliveira, Amâncio Jorge. (2006), "Eleições, política externa e integração regional". Revista de Sociologia e Política, pp. 145-155.
• Parlamento do Mercosul. (2021), "O Parlamento". Disponível em: https://www.parlamentomercosur.org/innovaportal/v/4300/2/parlasur/o-parlamento.html acesso em 05 de maio de 2021.
  » https://www.parlamentomercosur.org/innovaportal/v/4300/2/parlasur/o-parlamento.html
• Parlamento Europeu. “Fichas temáticas sobre a União Europeia: O Tratado de Lisboa”. Disponível em: https://www.europarl.europa.eu/factsheets/pt/sheet/5/o-tratado-de-lisboa acesso em 05 de maio de 2021
  » https://www.europarl.europa.eu/factsheets/pt/sheet/5/o-tratado-de-lisboa
• Pinheiro, Flávio. (2008), "Poder Legislativo e Política Externa: Um Estudo Sobre a Influência do Congresso Nacional do Chile na Formulação da Política Comercial durante a Década de 1990". Contexto Internacional, v. 30, pp. 89-139.
• Putnam, Robert. (1993), "Diplomacy and Domestic Politics: The Logic of Two-Level Games", in Evans, P., Jacobson, H., PUTNAM, R. (Eds.) International Bargaining and Domestic Politics - Double-Edged Diplomacy, University of California Press, Berkeley.
• Rabkin, Jeremy A. (2009), Law without nations?, Princeton University Press.
• Reimberg, Juliana; Andrade, Bruna Rodrigues. (2020), "Política Externa como Política Pública: Um Estudo Comparativo da Formulação da Política Externa no Brasil e no Uruguai". Anais X ANPAD - 2020
• Ribeiro, Pedro Feliú; Oliveira, Maria Letícia Claro. (2018), "Política Externa e Legislativo no Paraguai do Presidente Lugo". Revista de Sociologia e Política, v. 26, n. 68, pp. 75-94.
• Ribeiro, Pedro Feliú; Pinheiro, Flávio. (2016), "Presidents, legislators, and foreign policy in Latin America". Contexto Internacional, v. 38, pp. 467-501.
• Rihoux, Benoît; Ragin, Charles C. (2009), "Configural Comparative Methods. Qualitative Comparative Analysis (QCA) and Related Techniques". Applied Social Research Methods Series, v. 51. SAGE Publications Inc.
• Sandes-Freitas, Vítor; Bizzarro-Neto, Fernando. (2015). Qualitative Comparative Analysis (QCA): usos e aplicações do método. Revista Política Hoje, v. 24, n. 2, pp. 103-118.
• Santos, Manoel Leonardo; Cimini, Fernanda; Bohigues, Asbel. (2020), "Relações econômicas internacionais na América Latina: um estudo sobre as percepções das elites parlamentares". Opinião Pública, v. 26, n. 3, pp. 633-672.
• Scott, James M (1999). Public Opinion, Unpopular Policy and Congressional Activism: Exploring the "Mann Thesis" of US Foreign Policy-Making. Southeastern Political Review, v. 27, n. 4, pp. 679-697.
• Scharpf, Fritz. W. (2000). Gouverner l'Europe. Presses de Sciences Po.
• Souza, Matilde; Bahia, Bernardo Ramos. (2011), "Discussão e aprovação do Protocolo de Adesão da Venezuela ao Mercosul". Carta Internacional, v. 6, n. 1, pp. 67-82.
• Wolinetz, Steven B. (2006), "Party systems and party system types". Handbook of party politics, pp. 51-62.

Notes

• 1
  . Regionalization is a category of regionalism and is characterized by the movement of people and networks of communication, essentially culminating in a bottom-up dynamic (Hurrell, 1995; Börzel, Risse, 2016).
• 2
  . The Maastricht Treaty is available at: https://eur-lex.europa.eu/legal-content/PT/TXT/?uri=CELEX:11992M/TXT (Accessed on 10/05/2020).
• 3
  . It is important to note that this variable is significant in both systems but its effects differ. In parliamentary systems, it obstructs government formation and the stability of the head of government. In presidential systems, it impacts coalition composition and governability, given that the government isn’t shaped by the parliament and the president has a set term in office.
• 4
  . Twelve experts from the following institutions were consulted: Universidade de São Paulo (Brazil), Universidade Federal de Minas Gerais (Brazil), Universidade Estadual do Rio de Janeiro (Brazil), Universidade de Brasília (Brazil), Universidad Nacional de Colombia (Colombia), Fakultet političkih znanosti (Croatia), Aarhus Universitet (Denmark), Åbo Akademi (Finland), Università degli studi di Catania (Italy), Universidade de Lisboa (Portugal), University of Oxford (UK), and Universitatea Alexandru Ioan Cuza din Iași (Romania).
• 5
  . As mentioned in the explanatory model of the cases, the development of the survey with specialists aimed to select cases with a view to a comparative qualitative analysis. Respondents were asked to indicate cases of relevant events that occurred in Mercosur and the European Union that could serve to constitute an intentional sample for an empirical analysis of intermediate N. This is because, in the literature, the subject of legislative participation in international issues is controversial. Thus, it is possible to use the survey to reduce possible biases in our selection of cases.
• 6
  . We are aware that there is an unbalancing in case selection (2 from EU and 3 from Mercosur). However, the choice of QCA as an analysis method allows us – differently of quantitative techniques – to make an intentional selection that includes typical cases that reflect different configurations and results (see Sandes-Freitas, Bizzarro-Neto, 2013).
• 7
  . For further information regarding the classification of the variable for each decision, see the additional report available in Dados’ repository.
• 8
  . Available at: https://data.worldbank.org/
• 9
  . Despite ENP’s differing effects on the analyzed systems, this should not affect the inclusion of this causal condition in the model. Our hypothesis focuses on assessing if party fragmentation explains increased legislative participation in foreign policy decisions.
• 10
  . The main reason for choosing this form of measurement for the economic growth variable is that growth (GDP) varies annually but using only the previous year can generate a bias, as it may have been an atypical period. Thus, the average of the last four years mitigates this possible deviation.
• 11
  . Available at: https://data.worldbank.org/ (Accessed on 05/25/2021).
• 12
  . I.e.: participation is a function of presidentialisms AND bicameralism AND fragmentation AND prior debate AND growth OR commercial openness OR market concentration.
• 13
  . Consistency ranges from 0 to 1, where 1 represents perfect consistency. Consistencies above 0.9 are accepted.
• 14
  . The relevance of necessity (RoN) indicates how trivial a necessary condition is. This can occur in two ways: when the condition set is much larger than the result set of interest; or when the combination of the condition and the outcome is much greater than the absence of the outcome (Marisguia, 2020).
• 15
  . The sufficiency analysis undertaken will present complex solutions only for the four models. In section 4.3, an investigation will be carried out using CNA because this design provides a parsimonious solution equal to that of the QCA and the use of both forms of verification would lead to redundancy. We have chosen not to report the intermediate QCA solutions. Besides making the analysis very tedious, the pretest added little to the analysis. The results of all solutions can be consulted in the attached material, on Dataverse.
• 16
  . “an insufficient but necessary part of a condition which is itself unnecessary but sufficient for the result” (Mackie, 1965 apud Betarelli Jr., Ferreira, 2018:17).
• 17
  . This implies that, in the results of some CNA solutions, many causal chains can be identified - it is up to the researcher to select the solutions that best correspond to the empirical and theoretical framework.
• 18
  . A seminal article on this subject by Caporaso (1997).

Publication Dates

• Publication in this collection
  08 Nov 2024
• Date of issue
  Aug 2025

History

• Received
  7 June 2022
• Reviewed
  5 Nov 2022
• Reviewed
  24 Aug 2023
• Accepted
  24 Sept 2023