Fédéralisme Régionalisme

1374-3864 2034-6298


since 05 February 2011 :
View(s): 8177 (2 ULiège)
Download(s): 0 (0 ULiège)
Peter Bursens & Petra Meier

Belgian federalism 4.0: EU inspired ways forward

(Volume 20 : 2020 — Trente ans de dynamiques fédérales et régionales)
Open Access
Mots-clés : Belgium; European Union; supranationalism; intergovernmentalism; demoicracy; federalism


Over the last five decades Belgium has evolved into a federation with the aim of pacifying the divergent preferences of its constituent communities. In the light of the ongoing pressure on the current Belgian federal model, this article questions what other models might be appropriate for the Belgian political system. To that end, we take stock of three theoretical approaches used to describe the European Union, consider the extent to which they accord with the features of the Belgian federation, and discuss what remedies they might offer the Belgian case. The three approaches germane to the EU are: the intergovernmental or confederal model; the supranational or federal model; and the intermediary demoicratic or federation of states model. To this end, we pay specific attention to the relationship between the demos and the demoi, as well as between executive and legislative functions within and across government levels.

1. The future of the Belgian political system?1

1In the aftermath of the 2007 federal elections, De Winter and Baudewyns2 questioned whether the Belgian political model was still workable and discussed the possible breakdown of the Belgian nation-state. As Belgium had gone through a constitutional reform to the federal system five years before (in 2001-2), the lengthy negotiations to form a federal coalition government were puzzling. Their underlying argument was that the erstwhile fine-tuning ought to have implied a relatively smooth functioning of the Belgian political system with the federation now fully grown. Was the deadlock over government formation a symptom of a political system that had reached its expiration date? Could there be an alternative model after the unitary one, or the transitionary model that had transformed Belgium into a federal model?

2The authors did not really answer the question. The issue was mainly framed in terms of a potential breaking up of the country and nobody suspected that double the amount of time would elapse, namely 541 days, for a coalition to emerge in the wake of the 2010 federal elections three years later. Indeed, the negotiations continued whilst yet another constitutional reform (2011-2013), which amongst other things increased fiscal power at the regional level and decreased interregional solidarity3. After the elections of May 2019, nobody any longer expected the coalition negotiations to be quick. While consciously set aside by the Flemish nationalist party N-VA as the price of joining the federal government after the 2014 federal elections, the question of the state architecture once more found itself on the table during the electoral campaign, and in the months following the 2019 elections. The issue became even more topical in the heat of the Covid-19 pandemic, as the latter mercilessly uncovered the logistic costs the Belgian federal system prompts. Its complexity led to serious case specific deadlocks in matters of public health and security. Case specific deadlocks had already been apparent, for instance in the difficulties to adopt common positions in supranational and international negotiations on climate or trade (CETA) related policies. Their overall visibility during the pandemic but contributed to the general sentiment of a political deadlock. While some voices cautiously suggest that solutions to make the Belgian federation work more effectively should consider alternatives to a further downloading of competencies, such as a truly federal public health system and stronger cooperation mechanisms, the dominant tendency among Flemish political actors is to claim more autonomy in a fully-fledged confederation, if not complete independence.

3This contribution takes as its starting point the idea of a Belgian political system 4.0, to follow the unitary (1.0), transitionary (2.0) and federal (3.0) model defined by De Winter and Baudewyns4. As claims reach beyond a further fine-tune of the federal model, the question has become: what model next for Belgium? The objective of reflecting on a 4.0 model for Belgium is guided by more than the observation of overtly present deadlocks. Recently, Meier and Bursens5 assessed the Belgian federation regarding processes of representation and decision-making both within and between the government levels of the federation. They identified two problematic features. The first depicts the excessive dominance of the executives over the legislatures, putting constraints on the representative character of government within each of the government levels (constituent entities and federal level). They defined Belgium as a clear case of executive federalism, where the general tendency of executive dominance is strengthened by the way the federation and its rules and procedures were designed. The legislatures are left out of the picture, the Senate is powerless6, and the absence of state-wide parties turns party leaders into central brokers in Belgian politics7. These characteristics involve a domination of the executive both within and across government levels. The second problem points to the dominance of the multiple constituting demoi over the encompassing demos, diluting the representative character of the relations between government levels. The Belgian political system is not only characterized by the domination of the different – in the Belgian case – language groups over the encompassing people as such, but also by the almost complete lack of any formal recognition of the latter. Strong constitutional protection of the minorities, including language parity in governments, guaranteed representation of the language groups at the legislative levels, and double majority requirements8, go hand in hand with a weak bicameralism, characterized by a House of Representatives and other bilingual assemblies built upon the logic of representing and protecting the demoi, not the demos. In combination, again, with the absence of state-wide parties this impedes the accountability of governments to citizens9. In sum, the Belgian federation reflects a dominance of the executive over the legislative power and of the demoi over the encompassing demos.

4Such features of dominance make it interesting to use the EU as a source of inspiration when considering the future of Belgium, as the EU must also address the relation between the executive and the legislative bodies, and between the member states demoi and the (often contested) emergence of an overarching EU-wide demos. In considering what a Belgian political system 4.0 might look like if drawn from the theoretical approaches that underpin the multilevel polity of the EU, we first make the case that a comparison between Belgium and the EU is indeed an effective strategy. Secondly, we will present three distinctive approaches authors have developed for the EU (intergovernmentalism, supranationalism and demoicracy) and discuss the extent to which they can inspire the ongoing Belgian case. Finally, we outline the main findings alongside the limitations of our analysis and suggest avenues for further research.

2. The comparability of Belgium and the EU

5In considering what a Belgian political system 4.0 could look like, we draw inspiration from a comparison with the EU. The advantage of this is twofold. First, by looking at theoretical approaches applied to the EU, we remove the issue from the normatively and ideologically charged Belgian political dynamic. Secondly, options for the design of the multilevel polity of the EU have been thought through at numerous occasions, and, in that context, have been extensively scrutinized with respect to their representative credentials. In the Belgian case, the federalization process was never designed to ameliorate the existing form of representative democracy, nor did the architects of the federal state have a blueprint of the final product in mind10. While Belgium has been a parliamentary democracy ever since its creation in 1830, it first evolved into a consociational democracy11, only later into a federal system. The rationale behind both processes was to find constitutional and operational ways to govern the divided Belgian society. Consensual democracy was designed to overcome the socio-economic and religious cleavages, while federalism was developed to appease the ethno-linguistic division. It is therefore not surprising that the effect of the federalization process on the democratic functioning of the Belgian state has seldom been examined12, while such types of analysis have been conducted with respect to the EU13.

6To draw inspiration from the EU assumes that both political systems are comparable. Multiple authors have argued that the EU and (quasi) federal states are sufficiently alike to allow for fruitful comparisons. This is obvious for those who regard the EU as a politicized multilevel polity14. Some authors15 even argue that the EU can be defined as a highly institutionalized and hierarchical international organization. Of course, the EU and Belgium have been subject to different constitutional and institutional dynamics. On the one hand, the EU can be regarded as a case of centripetal nation building, where integration has long been an almost exclusively elite-driven process. For a long time, neither EU institutions nor policies were politicized. It is only recently that contestation by segments of the electorate has emerged, triggering demands for the politicization of EU policies and even the polity itself16. Belgium, on the other hand, is characterized by a centrifugal dynamic17. Unlike in the EU, the falling apart has initially taken place in a very politicized way, certainly at the level of political parties, but also at the level of the electorate18. However, while over the course of time the issue remained salient for a large part of the Flemish political elite, public opinion did not become more politicized on the issue19. Public contestation of the overall polity is less the case in Belgium than in the EU.

7In their own way, both European and Belgian institutional reforms have been subject to path dependent pressures. Regarding the EU, Bevir and Phillips20 argue that the consecutive treaty reforms can be defined as system maintenance, meaning that the options for democratic reforms in the EU were limited to those that fit the centripetal integration logic. In this respect, Scharpf21 argued that the EU is liberal but not republican. Institutional reform by majority voting is not accepted by the constituent member states as the EU lacks a collective identity and public sphere. Therefore, the Treaty prescribes the tough requirement of unanimity to change the institutional architecture, ensuring the outcome is almost certain to be conservative. A similar reasoning can be applied to the Belgian case where democratic reforms in constitutional revisions have been limited to what fits the centrifugal logic22: limited attention for the federal level and a Belgian demos, and a development and strengthening of constituent entities with strong executives that dominate intergovernmental relations, including those that lead to state reform agreements. Unlike the EU, however, constitutional reform must only be supported by a majority (albeit a special double majority – demanding an overall two-third majority plus a simple majority in each language group), despite the lack of a state-wide public sphere, mass media or parties.

8The next section will discuss three dominant theoretical approaches on the EU23: intergovernmentalism; supranationalism; and demoicracy. It goes without saying that the EU does not fully match with any of them, and neither do we expect the Belgian case to do. However, this is not important as the main aim of this contribution consists in finding inspiration in these coherent models for the Belgian case. We critically analyze these approaches to consider what possible lessons can be drawn for the Belgian case when it comes to the imbalance between the encompassing demos and constituting demoi, as well as the imbalance between the legislative and executive bodies.

3. EU models applied to the Belgian case

3.1. The intergovernmental model

9The oldest theoretical approach on the EU stems from an international relations view of European integration. Hoffmann’s intergovernmentalism24 defines the EU as an international organization, i.e. as an organization set up by sovereign states and one that has a limited functional scope, seeking solutions that serve all participating states. Its output does not entail winners or losers, redistribution, nor any top down enforcement backed by a supranational court or dispute settlement mechanism. In the same vein, Majone25 argues that the regulatory character of most EU policies do not call for majoritarian institutions at the EU level. Technocratic rule is seen as enough or even necessary to reach pareto-optimal policies that benefit all involved. The more advanced perspective of liberal intergovernmentalism of Moravcsik26 adjusted the original theory by stressing that the dynamics within member states should be considered to understand the bargaining positions of the states constituting the international organization. The outcome at the international level, however, is still seen as the result of international negotiation among the representatives of the states’ executives. This is the essence of what Putnam27 has called the two-level game. From a democratic perspective, this entails that the international level, i.e. the EU level, should not in itself be evaluated in democratic terms. As long as the member states function as fully developed democracies, with well-functioning national channels of democratic representation, the EU can be regarded as taking legitimate decisions.

10Intergovernmentalism applied at the level of a state would mean a confederal political system. Applying the intergovernmental approach to the Belgian case points to a radical way forward: breaking up the Belgian state into a set of independent states, which, as representative democracies, decide on a series of common policies to be tackled at a newly constituted confederal level. In addition, these collective decisions should be taken unanimously as this would allow the governments of each constituent entity to remain fully accountable to their respective demoi.

11Such an intergovernmental approach would solve the dominance of the constituent demoi over the encompassing demos by making the level of the Belgian demos redundant. Confederations do not conceptualize an overarching demos. Indeed, this model would imply that no representative democracy, and hence no parliament, is installed at the confederal level, only an executive body that decides by unanimity. The model would make the Senate redundant, as well as all the rules regulating the relations between the language groups. Such an intergovernmental model actually reflects some of the positions of both the Flemish nationalist Nieuw-Vlaamse Alliantie (N-VA) and the Flemish radical right Vlaams Belang (VB), and to a lesser extent also of the Flemish Christian Democrats (CV&V). Especially the first two emphasize the existence of two democracies, by which they understand the two major language groups, the Flemish and the French speaking population. Speaking of two democracies, however, does not take into consideration the existence of all the demoi in Belgium, of all the communities and regions and their complex relations. It provides no option for the Brussels Capital Region, which is bilingual, neither does it offer much for the German-speaking Community, which is not bound to territory, both of which are socio-demographic realities. It also clashes with a model generally put forward by the German-speaking Community, thinking of Belgium in terms of four regions rather than two communities28.

12Also, a confederal design would not necessarily outbalance the dominance of the executive in the Belgian case. Confederal relations are by definition intergovernmental. In that respect such a model would not make much of a difference as compared to the current situation, in which relations between the federal level and the constituent entities or among constituent entities are completely in the hands of the executives (cooperation agreements, negotiations on a common Belgian position at the supranational or international level, prevention or solution of conflicts of interest or competencies, etc). Nonetheless, there still is a federal parliament, controlling the federal government, whereas in a confederal model only the parliaments of the constituent entities control their governments. Also, the dominance of the executive over the legislative would depend on the way the parliaments of the constituent entities control their executives, which may differ from one constituent entity to another in a confederation.

13Finally, decision-making by unanimity at the confederal level means a high probability of non-decisions and deadlock, which will most likely result in a limited number of confederal policies, actually only those where cooperation is deemed more important than the risk of deadlock. Referring to Scharpf’s29 distinction between input and output legitimacy, the intergovernmental option favors legitimacy at the input side (as fully representative governments can veto common policies), but risks to fail at the output side (as the unanimity requirement may result in a low propensity of problem-solving capacity due to non-decisions and deadlock). In this respect, a confederal model does not offer the Belgian case a lot to overcome its current problems of deadlock. On the contrary, the current context in which the same parties serve at the federal level and at that of the constituent entities forces them to coordinate their activities across levels and thus to consider the stakes at play at these different levels30. And the federal parliament tends to function, especially in phases of a caretaker government31. These mechanisms would disappear in a confederal model without being replaced by anything else.

3.2. The demoicratic model

14A second approach that defines the EU polity is related to the federation of states model used to depict national models32 and situates itself between the intergovernmental and supranational approach. It has been labelled demoicracy33, a ‘polity of democratic states with hierarchical, majoritarian features of policy-making, especially in value-laden redistributive and coercive policy areas, but without a unified political community (demos)’34. In other words, demoicracy takes the basic intergovernmental position of the separate political communities of states as building blocks of the polity but combines them with joint competences on a common level that are subject to majoritarian decision-making. In its vertical dimension, demoicracy is therefore based on the equality and interaction of individual citizens’ (European demos) and member states’ (national demoi) representatives in the making of common policies. Horizontally, it seeks to balance equal transnational rights of citizens with national policy-making autonomy.

15Cheneval and Schimmelfennig35 argue that the EU should be regarded as a demoicracy rather than a supranational or intergovernmental system. Citizens in the EU rarely identify themselves with the European level polity, the EU does not have an integrated public sphere nor well-developed transnational mass media, and only weak intermediary structures with which to mobilize political preferences (there are no EU-wide political parties, only loosely organized party federations). Furthermore, a set of basic principles strike a balance between the rights of the European demos and the rights of the national demoi and allow for evaluating the democratic character of the polity. First, demoi should be free to join and to leave the encompassing union, should be able to veto new treaties that establish the common rules, and should be democracies themselves. Secondly, there should be no discrimination of demoi and demos. Thirdly, demoi and demos should have equal legislative rights in the encompassing polity. Finally, the laws of the union should have primacy over the laws of the constituent entities and a supreme court should have jurisdiction over all dimensions of the polity that have been attributed to the polity via the treaty.

16Looking at the Belgian system through a demoicracy lens, much like in the case of the intergovernmental approach, departs from the notion of sovereign states as constituent entities of the encompassing polity. However, demoicracy points to a common level with joint competences subject to majoritarian decision-making. This requires a form of bicameral system through which the demoi and the demos are both represented at the federal level and in which both enjoy an equal position when it comes to federal law-making. In other words, in a demoicracy the demoi and the demos should participate on equal footing in federal law-making. In the current Belgian system, however, there is no full bicameralism whereby two chambers are equally powerful in law-making. Only the House of Representatives (Kamer/Chambre) has law-making competences. The Senate (Senaat/Sénat) hardly plays any significant role in law-making36. However, members of the House of Representatives are elected in separate constituencies that coincide with the demoi. In the House of Representatives, the demoi, rather than the Belgian demos, are represented, which is further strengthened by the logic of language groups and minority protecting rights. The Senate is, of course, the assembly where the demoi are even more clearly and formally represented. In sum, the demoicracy model would call for the return to a fully-fledged bicameral system with the House of Representatives representing the Belgian demos and the Senate the demoi, and both playing a significant role in law-making.

17We also need to consider the horizontal dimension which calls for equal rights of citizens in common decision-making. In practice, this comes down to avoiding negative consequences of decisions taken by one constituent entity for one or several others. This avoidance of negative externalities is hindered in Belgium because executive or legislative actors lack incentives to care about such spill-over effects. Parties only need to receive votes within one language group. Once elected and in government, they are neither accountable to members of other language groups for the negative externalities that result from decisions taken in, and by, the language group by whom they have been elected, nor for decisions taken at the federal level. The quasi absence of state-wide parties and of a state-wide constituency is crucial here. This is obvious in cases where one constituent government takes decisions that are considered harmful by another. It may also be relevant in the case of vertical negative externalities. For example, a federal minister belonging to a French-speaking party need not fear electoral backlash for a decision that may harm the Flemish government, and vice versa.

18Belgium does have a constitutional court that acts as a supreme court, interprets the constitution, and guards over the correct application of the division of competences between government levels. However, unlike in the EU polity, Belgium has no courts at levels other than the federal one. A further significant difference between the EU and Belgium concerns the hierarchical relation between legal norms as the supremacy of federal law over constituent entity law is absent in Belgium. Put differently, the demos cannot overrule the demoi, which makes Belgium an extreme example of the separation between government levels.

19The demoicracy approach also highlights the dominance of the executive over the legislature with respect to Belgium. Based on the principle of a fully-fledged bicameral system, both demoi and demos need to be represented at the supranational level and participate on an equal footing in law-making. Whilst the EU does indeed have a bicameral system that honors the equal representation of the demos and the demoi, it is strongly arguable that executive actors dominate legislative actors. Despite acting as a legislator, the Council for example, is composed of executive actors who represent member state governments. The remedies often proposed at the EU level are to increase the power of the EP, and above all, to call for more scrutiny of EU policies by national parliaments. Equally, cooperation between (national and European) parliaments and members of parliament may give legislative actors more leverage in their relationship with executive actors.

20This perspective on the EU is difficult to apply to Belgium on the basis that there is no balanced bicameral system. Strengthening the House of Representatives vis-à-vis the federal executive may enhance parliamentary powers at the federal level, however, this does not remedy the imbalance between the demos and the demoi. Stimulating cooperation between parliaments, both horizontally within one level and vertically across levels is unlikely to have direct consequences either because each parliament has a relationship with its own executive within its jurisdiction only. Of course, cooperation can lead to effects in terms of learning from best practice, but not necessarily to greater leverage for the federal parliament over the federal executive. More involvement of constituent parliaments in the positions that constituent executives take in federal matters would be rendered irrelevant because of the dual character of Belgian federalism. Constituent executives, and parliaments, for that matter, are constitutionally excluded from federal law-making. In Belgium, levels of government are only linked informally through the party leaderships of the parliamentary majorities at the various levels. In sum, introducing reforms to strengthen the legislature vis-à-vis the executive along the lines of the EU demoicracy model is likely to have little effect if not accompanied by reforms balancing the relation between the demoi and the demos.

3.3. The supranational model

21The third model conceptualizes the EU as a supranational political system. This model shares with the demoicracy approach an emphasis on common decision-making that respects both individual citizens and the constituent polities. However, in contrast to the demoicracy (and the intergovernmental) approach, it assumes that the encompassing polity reflects a political community.

22Føllesdal and Hix37 propose to reform the EU to achieve a genuine politicization of EU-level politics. What they propose is a true contestation for EU level policy alternatives and political elites, made possible by procedures such as elections on an EU-wide scale. Such reforms would also alter the relation between the European demos and the national demoi. In the current system, the demoi are dominant because representation at the EU-level rests upon national political communities, national political parties, and national electoral constituencies. By turning the electoral process into a truly EU-wide event, the encompassing European demos would be more in balance with the national demoi. Of course, this either assumes that a single European demos is latently present, or that such a demos can be developed via institutional engineering. It is also argued that such changes would make the EU-level executive fully accountable to an EU-level legislature. In this scenario, the European executive, currently the European Commission, would have to be transformed into an EU government that rests upon a majority in the EP, becoming the second chamber in a European bicameral system (the other chamber being the current Council of Ministers). Such a reform can also affect the relationship between the executive and the legislative bodies. The EP would not only have the power to approve and impeach the executive as is the case now but would also be able to exert more influence over the policies proposed by the executive. Additionally, the EP could also be granted the right to initiate legislation. As an alternative to the parliamentary system, one could think of a more presidential type of system in which the president of the European Commission would be directly elected by an EU-wide constituency. Finke38 argues that parties would then become more important at the EU level. MEPs who are instructed by their national governments on how to vote, have a greater propensity to defect from their European party groups. Parties can be the vehicles to coordinate between the EP and the Council, but this seems only to be the case in strong domestic parliaments (such as Denmark or Germany).

23At first sight, the Belgian case matches the supranational model far better than the EU, because the Belgian federal executive rests upon a majority in the second chamber (the House of Representatives) of the federal parliament. However, taking a more detailed look, some remarkable similarities arise between the European and the Belgian system. The way members of the federal parliament are elected is similar to the European process: there are no state-wide parties and there is no state-wide electoral college. This makes the consequences for responsiveness similar to the EU as well: voters can choose rival policies and policymakers at the federal level, but not for all the parties or candidates who may end up in parliament or government. In addition, voters cannot kick out the executive actors they have not been able to vote for, although all executive actors can take decisions that affect them. With respect to political parties one could even argue that the EU is far more integrated than Belgium because of the existence of EU-wide party federations, and more so because of the multinational political groups in the EP. Party federations and joint parliamentary groups do not exist in the Belgian system, except for the radical left and the Greens.

24Reforming the EU from a supranational approach comes down to turning the EU into a cooperative federal system, be it parliamentary or presidential. Either the politicization at the European level is organized via EU-wide elections for the EP from which a majority emerges that supports the EU executive, or it is organized via EU-wide elections designed to directly elect the president of the executive. At the EU level, several additional remedies can be envisaged, such as the enhanced scrutiny of domestic parliaments on the EU policies of national governments, a strengthened position of national parliaments as multi-arena players, or an increased use of all sorts of formal and informal interparliamentary cooperation tools to strengthen the position of the legislatives vis-à-vis the executives39.

25Translated to the Belgian case, the supranational approach is a call to replace the current dual federal system with a cooperative one, and more specifically a type of parliamentary cooperative federalism. In the Belgian case cooperation does exist wherever EU policies are situated at the constituent level, but these forms of cooperation mainly take place among executives, with the federal government as a coordinator40. A parliamentary cooperative federalism would entail several elements. First, it would involve a bicameral system with a House of Representatives representing the demos and a Senate representing the demoi. Secondly, it would imply the politicization of the House of Representatives via a state-wide electoral constituency involving state-wide political parties. Such a system would restore the balance between the encompassing Belgian demos and the constituting demoi. Thirdly, it would require installing interparliamentary cooperation, which is now almost non-existent, to strengthen the position of the legislative vis-à-vis executive actors. Such cooperation can be both horizontal and vertical. Fourthly, and again parallel to the EU, regional and community parliaments could become involved in constitutional reform negotiations, instead of the current position of party leaders dominating in these matters. Finally, like the EU level, political parties can take up a coordinating role between government levels and institutions in Belgium. While this coordination is currently done by the leadership of political parties, and basically takes place outside the parliamentary arena, a transfer from party headquarters to the parliamentary venue would strengthen the position of the legislatures vis-à-vis the executives.

4. Conclusions

26This article speaks to the ongoing discussion about how the Belgian political system should evolve. Drawing on three theoretical approaches commonly put forward for the reform of the EU, namely intergovernmentalism, supranationalism and demoicracy, we considered their merits as political models for the Belgian federation. We paid specific attention to the relation between the Belgian demos and the constituting demoi as well as to the position of the executive vis-à-vis legislative bodies. Both are critical features of the Belgian system and both have been extensively reflected upon for the EU polity. To what extent then would these approaches, designed for the multilevel polity of the EU, suit a Belgian political system 4.0? Formally being a federation, taking inspiration from the supranational model is the most obvious way to proceed. However, we also showed how both demoicracy and intergovernmentalism share certain features with the current Belgian model.

27The intergovernmental approach implies a Belgian confederation composed of separate demoi in sovereign states which enjoy a veto-power at the confederal level. The imbalance between the demoi and the demos would become irrelevant as this model considers an encompassing Belgian demos to be non-existent. If the constituting states restore the balance between their respective legislatures and executives at the same time, one could argue that such a model would indeed increase input legitimacy of the confederation. However, it would also put major constraints on output legitimacy as the risk for deadlock and non-decisions would increase due to the requirement for unanimous decision making at the confederal level. In addition, the feasibility of this model raises questions as it implies deleting the current federal state with the consent of all constituent entities (and thus solving the Brussels conundrum and providing space for the German-speaking Community) and rebooting it as a confederal state.

28Implementing the demoicracy model would require an encompassing level with equal representation of the Belgian demos and constituent demoi via a bicameral system. Both should be equally involved in majoritarian decision-making for Belgian-level legislation. This would mean substantial changes to the bicameral system, the electoral constituencies and it would require increased within and across level parliamentary scrutiny on all executive bodies. The demoicracy model differs from the intergovernmental one as it allows for majoritarian decision-making at the common level but shares the assumption that a common political community is absent. The feasibility concerns, namely the dissolving and subsequent installation of a common Belgian level are similar as well.

29The supranational approach assumes the existence of both an encompassing political community and separate constituent political communities. The current system would need quite a radical shift from dual to cooperative federalism with a bicameral legislative that is composed of a state-wide elected House of Representatives and a Senate that represents the constituent entities. Different to the demoicratic model is the emphasis on politicization (defined as competition for political leadership and policies) at the federal level through a state-wide electoral constituency via state-wide political parties. The latter would provide the House of Representatives with a genuine demos-based membership in balance with the demoi-based membership of the Senate. An increased scrutiny role for the parliaments, both within and across levels, would additionally help to restore the legislative – executive balance. Federal reform in this direction is also prone to feasibility concerns as many of the described requirements would imply swimming against the tide of former decisions, and hence challenging the powerful logic of path dependency.

30While technically speaking the theoretical approaches to grasp the multilevel polity of the EU each contain valuable sources of inspiration for the further reform of the Belgian political system, either in terms of improving the current federal model or in terms of moving to a new one, the key question that arises is one of political feasibility. The current Belgian federation showcases characteristics of all three models set out for the multilevel polity of the EU. This reflects the – subjectively experienced or objectively given – need for incremental short-term political compromises and the incapacity to make a clear-cut choice for a coherent political model. Having discussed three models designed for the multilevel polity of the EU does not change that, but it does contribute to a clearer understanding of what a coherent, and in that respect balanced, political model might look like. Clearly, our analysis was confined to a limited set of institutional features related to democratic representation. Considering other dimensions such as the electoral system and landscape of political parties, the multilevel organization of the judiciary, the supremacy of some legal acts over others, or the implementation stages of the policy cycle may further nuance our findings. Alternatively, comparisons with other federal systems – and their evolution – might help finding additional ways to further develop the Belgian political system.


1 We would like to thank Dirk De Bièvre, François Randour, and the anonymous reviewers for their rich and thoughtful comments on an earlier version of this article.

2 De Winter (L.) and Baudewyns (P.), ‘Belgium: Towards the Breakdown of the Nation-State in the Heart of Europe’, Nationalism and Ethnic Politics, vol. 15, n° 3-4, 2009, pp. 280-304.

3 Popelier (P.), Sinardet (D.), Velaers (J.) and Cantillon (B.) (eds.), België, quo vadis? Waarheen na de zesde staatshervorming?, Antwerpen, Intersentia, 2012.

4 De Winter (L.) and Baudewyns (P.), ‘Belgium: Towards the Breakdown of the Nation-State in the Heart of Europe’, Nationalism and Ethnic Politics, vol. 15, n° 3-4, 2009, pp. 280-304.

5 Meier (P.) and Bursens (P.), ‘Belgium: the Democratic State of the Federation’, in Benz (A.) and Sonnicksen (J.) (eds.), Federal Democracies at Work: Varieties of Complex Government, Toronto, University of Toronto Press, 2021, pp. 180-196.

6 Dandoy (R.), Dodeigne (J.), Reuchamps (M.) and Vandeleene (A.), ‘The new Belgian Senate. A (dis)continued evolution of federalism in Belgium?’, Representation, vol. 51, n° 3, 2015, pp. 327-339.

7 De Winter (L.), Swyngedouw (M.) and Dumont (P.), ‘Party System(s) and Electoral Behaviour in Belgium: From Stability to Balkanisation’, West European Politics, vol. 29, n° 5, 2006, pp. 933-956.

8 Pilet (J.-B.) and Pauwels (T.), ‘De vertegenwoordiging van taalgroepen in België : de weg naar een hyperinstitutionalisering’, in Celis (K.), Meier, (P.) and Wauters (B.) (eds.), Gezien, gehoord, vertegenwoordigd? Diversiteit in de Belgische politiek, Ghand, Academia Press, 2010, pp. 47-68.

9 Deschouwer (K.), ‘And the peace goes on? Consociational democracy and Belgian politics in the twenty-first century’, West European Politics, vol. 29, n° 5, 2006, pp. 895-911.

10 Deschouwer (K.), The Politics of Belgium. Governing a Divided Society, Basingstoke, Palgrave, 2012.

11 As defined by Lijphart (A.), Democracy in Plural Societies: A Comparative Exploration, New Haven, Yale University Press, 1977. Lijphart (A.), Patterns of Democracy: Government Forms & Performance in Thirty-six Countries, New Haven, Yale University Press, 2012.

12 Amongst the exceptions are De Grauwe (P.) and Van Parijs (P.) (eds.), Electoral Engineering for a Stalled Federation. A country-wide electoral district for Belgium’s federal Parliament, Brussels, Re-Bel E-book, 2009, disponible sur http://www.rethinkingbelgium.eu/rebel-initiative-files/ebooks/ebook-4/Re-Bel-e-book-4.pdf (consulted 20/09/2019). De Grauwe (P.) and Van Parijs (P.) (eds.), The Malaise of Electoral Democracy and What to Do About It, Brussels, Re-Bel E-book, 2014, disponible sur http://www.rethinkingbelgium.eu/rebel-initiative-files/ebooks/ebook-14/Re-Bel-e-book-14.pdf (consulted 20/09/2019). van Haute (E.) and Deschouwer (K.), ‘Federal reform and the quality of representation in Belgium’, West European Politics, vol. 41, n° 3, 2018, pp. 683-702.

13 Fossum (J. E.) and Jachtenfuchs (M.) (eds.), ‘Federal challenges and challenges to federalism. Insights from the EU and federal states’, Journal of European Public Policy, vol. 24, n° 4, 2017, pp. 467-632.

14 Fabbrini (S.), Federalism and Democracy in the European Union and the United States, Londres, Routledge, 2005. Fabbrini (S.), Compound Democracies. Why the European Union and the United States are becoming Similar, Oxford, Oxford University Press, 2012. Hix (S.), ‘The EU as a Polity’, in Jørgensen (K.), Pollack (M.) and Rosamunde (B.) (eds.), The Sage Handbook of European Union Politics, New York, Sage, 2006, pp. 141-158. Hooghe (L.) and Marks (G.), Multilevel Governance and European Integration, Lanham, Rowman and Littlefield, 2001. Hooghe (L.) and Marks (G.), ‘A Postfunctionalist Theory of European Integration: From Permissive Consensus to Constraining Dissensus’, British Journal of Political Science, vol. 39, n° 1, 2009, pp. 1-23. Nicolaïdis (K.) and Howse (R.), The Federal Vision, Oxford, Oxford University Press, 2001.

15 Farrell (H.) and Newman (A.), ‘The New Interdependence Approach: Theoretical Development and Empirical Demonstration’, Review of International Political Economy, vol. 23, n° 5, 2016, pp. 713-736.

16 McNamara (K.), ‘Authority Under Construction: The European Union in Comparative Political Perspective’, Journal of Common Market Studies, vol. 56, n° 7, 2018, pp. 1510-1525. Olsen (J. P.), ‘Democratic accountability and the changing European political order’, European Law Journal, vol. 24, n° 1, 2018, pp. 77-98.

17 Swenden (W.) and Jans (M. T.), ‘‘Will it stay or will it go?’ Federalism and the sustainability of Belgium’, West European Politics, vol. 29, n° 5, 2006, pp. 877-894.

18 Luyten (D.), Meynen (A.) and Witte (E.), Politieke geschiedenis van België van 1830 tot heden, Antwerpen, Manteau, 2016.

19 Reuchamps (M.), Sinardet (D.), Dodeigne (J.) and Caluwaerts (D.) ‘Reforming Belgium’s Federalism: Comparing the Views of MPs and Voters’, Government & Opposition, vol. 52, n° 3, 2017, pp. 460-482.

20 Bevir (M.) and Phillips (R.), ‘EU Democracy and the Treaty of Lisbon’, Comparative European Politics, vol. 15, n° 5, 2017, pp. 705-725.

21 Scharpf (F. W.), ‘Legitimacy in the European Multilevel Polity’, European Political Science Review, vol. 1, n° 2, 2009, pp. 173-204.

22 Deschouwer (K.), The Politics of Belgium. Governing a Divided Society, Basingstoke, Palgrave, 2012.

23 Nicolaïdis (K.), ‘Our European demoi-cracy: is this constitution a third way for Europe?’, in Nicolaïdis (K.) and Weatherill (S.), eds., Whose Europe? National Models and the Constitution of the European Union, Oxford, Oxford University Press, 2003, pp. 137-152.

24 Hoffmann (S.), ‘Obstinate or Obsolete? The Fate of the Nation-State and the Case of Western Europe’, Daedalus, vol. 95, n° 3, 1966, pp. 862-915.

25 Majone (G.), ‘The Regulatory State and its Legitimacy Problems’, West European Politics, vol. 22, n° 1, 1999, pp. 1-24.

26 Moravcsik (A.), The Choice for Europe, Ithaca, Cornell University Press, 1998.

27 Putnam (R. D.), ‘Diplomacy and Domestic Politics. The Logic of Two-Level Games’, International Organization, vol. 42, n° 3, 1988, pp. 427-460.

28 Van Parijs (P.), Belgium. Une utopie pour notre temps, Brussels, Transversales, 2018.

29 Scharpf (F. W.), ‘Legitimacy in the European Multilevel Polity’, European Political Science Review, vol. 1, n° 2, 2009, pp. 173-204.

30 Pilet (J.-B.) and Meier (P.), ‘Ze halen hun slag wel thuis. Over particratie en het aanpassingsvermogen van Belgische partijen’, Res Publica, vol. 60, n° 4, 2018, pp. 321-345.

31 Van Aelst (P.) and Louwerse (T.), (Parliament without Government. The Belgian Parliament and the Government Formation Process of 2007-2011’, West European Politics, vol. 37, n° 3, 2014, pp. 475-496.

32 Rummens (S.) and Sottiaux (S.), ‘Democratic Legitimacy in the Bund or ‘Federation of States: the Cases of Belgium and the EU’, European Law Journal, vol. 20, n° 4, 2014, pp. 568-587.

33 Cheneval (F.) and Schimmelfennig (F.), ‘The Case for Demoicracy in the European Union’, Journal of Common Market Studies, vol. 52, n° 2, 2013, p. 334-350. Nicolaïdis (K.), ‘European Demoicracy and its Crisis’, Journal of Common Market Studies, vol. 51, n° 2, 2013, pp. 351-369.

34 Cheneval (F.), Lavenex (S.) and Schimmelfennig (F.), ‘Demoi-cracy in the European Union: Principles, Institutions, Policies’, Journal of European Public Policy, vol. 22, n° 1, 2015, pp. 1-18 – citation p. 353.

35 Cheneval (F.) and Schimmelfennig (F.), ‘The Case for Demoicracy in the European Union’, Journal of Common Market Studies, vol. 52, n° 2, 2013, pp. 334-350.

36 Dandoy (R.), Dodeigne (J.), Reuchamps (M.) and Vandeleene (A.), ‘The new Belgian Senate. A (dis)continued evolution of federalism in Belgium?’, Representation, vol. 51, n° 3, 2015, pp. 327-339.

37 Føllesdal (A.) and Hix (S.), ‘Why There is a Democratic Deficit in the EU: A Response to Majone and Moravcsik’, Journal of Common Market Studies, vol. 44, n° 3, 2006, pp. 33-62.

38 Finke (D.), ‘Bicameralism in the European Union: Parliamentary Scrutiny as a Tool for Reinforcing Party Unity’, West European Politics, vol. 40, n° 2, 2017, pp. 275-294.

39 Auel (K.) and Neuhold (C.), ‘Multi-arena players in the making? Conceptualizing the role of national parliaments since the Lisbon Treaty’, Journal of European Public Policy, vol. 24., n° 10, 2017, pp. 1547-1561. Fromage (D.), ‘Increasing Inter-Parliamentary Cooperation in the European Union: Current Trends and Challenges’, European Public Law, vol. 22, n° 4, 2016, pp. 749-772. Gattermann (K.) and Hefftler (C.), ‘Beyond Institutional Capacity: Political Motivation and Parliamentary Behaviour in the Early Warning System’, West European Politics, vol. 38, n°. 2, 2015, pp. 305-334.

40 Beyers (J.) and Bursens (P.), ‘How Europe Shapes the Nature of the Belgian Federation: Differentiated EU Impact Triggers Both Cooperation and Decentralization’, Regional and Federal Studies, vol. 23, n° 3, 2013, pp. 271-291. Happaerts (S.), Schunz (S.) and Bruyninckx (H.), ‘Federalism and Intergovernmental Relations: The Multi-Level Politics of Climate Change Policy in Belgium’, Journal of Contemporary European Studies, vol. 20, n° 4, 2012, pp. 441-458.

To cite this article

Peter Bursens & Petra Meier, «Belgian federalism 4.0: EU inspired ways forward», Fédéralisme Régionalisme [En ligne], Volume 20 : 2020, Trente ans de dynamiques fédérales et régionales, URL : https://popups.uliege.be/1374-3864/index.php?id=2050.

About: Peter Bursens

Full Professor in Politics, Department of Political Science and GOVTRUST Centre of Excellence, University of Antwerp

About: Petra Meier

Professor in Politics, Department of Political Science and GOVTRUST Centre of Excellence, University of Antwerp