1 Very Very Preliminary Draft IPSA 24 th World Congress of Political Science Poznan July 2016 The vulnerability of the European Elite System under a prolonged crisis Maurizio Cotta (CIRCaP- University of Siena)
2 The vulnerability of the European Elite System under a prolonged crisis Maurizio Cotta (CIRCaP- University of Siena) In the years after 2008 a prolonged economic crisis has affected the European countries most of which are members of the EU. On top of this crisis multiple security problems on the Eastern and Southern borders of the EU and a severe immigration crisis have generated new yet unexplored challenges for the European Union. Which have been the consequences of all this for the European Union? The political impact of economic downturns has been often discussed at the national level especially in the context of electoral studies, which have thoroughly analyzed the influence of economic factors upon voting behaviour (Bellucci and Lewis Beck). Overall the hypothesis that economic bad times will negatively affect the incumbent parties and leaders is broadly confirmed (Lewis Beck and Nadeau 2011). This does not exclude however the mediating role of other variables. The impact of economic factors at a more aggregate level such as the party system has received less systematic attention and findings are not particularly encouraging (Tavits 2006, Tavits 2008). The impact of other crises (such as immigration) has received increasing attention in recent times (Oesch 2008) but we are still far from having well established theories. Quite understandably given the predominant state-centric perspective adopted in political science the impact of economic and other crises has normally been discussed with regards to national political systems and their workings. In this paper the focus
3 will rather be on that compound (Cotta 2012, 2014, Fabbrini 2007) or multilevel (Hooghe and Marks 2001) polity which is the EU. The purpose of this paper is predominantly explorative and descriptive. A more advanced explanatory perspective will be developed in the next version of this paper. The European Elites System (EES) is the central concept that guides my analysis. Let me briefly specify the meaning of this concept. Given the peculiar structure of the European polity its political elites have also a particularly complicated structure. In fact we should talk of an Elite System made of a plurality of different components; I will call this the European Elite System. Which are its components? The first component of the EES is the multiple set of national elites of the member states (28 today until the Brexit will be implemented; then 27) which in different ways play not only a national role but also a European one (whenever their national decisions impinge upon European decisions). This component is also directly responsible for the production of the second component of the EES: the members of the European Council and of the Council of the European Union (formerly the Council of ministers). As it is well known the two institutions are composed by the incumbent members of the corresponding national institutions. This component has therefore an amphibious nature as it plays, at the same time, top level and very demanding national and European roles. A similar position have the Governors of the national banks who are part of the Council of the ECB while maintaining their position at home. The important feature of this elite component is that the European institutional role of its members lasts only as long as their national institutional positions are continued. The third component is made of
4 those who work permanently in the central institutions of the Union: political members and high bureaucrats of the European Commission; members of the European Parliament; members of the Executive Board of the European central bank, and judges of the High Court of justice. The members of this third elite component have important national roots and linkages as their recruitment into the supranational institutions is strictly connected to their national identity, however their institutional affiliation is linked only to one of the European institutions. Because of its composite nature the state of the EES is bound to be affected by a plurality of influences. In order to study the impact of the crisis the different components must be treated first separately as their life depends from different institutional mechanisms. Then it must be discussed what have been the systemic outcomes of these changes. The broad question that guides this paper is to assess to what extent the vulnerability of the EES has increased in the crisis period. By vulnerability I mean the probability of elite members to lose their position in the institutional circle of power. The national level effects In view of the fact that the most effective democratic mechanisms of reaction (representation and accountability) to the problems and demands raised from the population(s) of the European Union are still national, I will start from the nation level effects and then move to European indirect and direct effects. To assess the
5 vulnerability of national political elites I will concentrate my attention upon the consequences of elections upon national parliaments and cabinets and compare the crisis period ( ) with a pre-crisis period ( ) of similar length. With regard to parliamentary elites the first indicator I will use is the volatility index and check to what extent it has increased over time (at a later stage of this research I will also analyse the parliamentary turnover rate and the rise of new parties). The second indicator I will then use is the cabinet incumbency effect, that is to say the amount of electoral losses or gains obtained by the incumbent government parties at each election. I will then analyse the prime ministerial vulnerability, i.e. the frequency of prime minister changes after an election. To these I will add later (in the next version of the paper) the majority alternation index, i.e. the probability of a total or partial change in the government majority after an election. The volatility index (or Pedersen index) has been frequently used to evaluate the degree of stability and institutionalisation of party systems. My expectations are that in a crisis period voters should be more inclined to defect (especially from governing or more in general from established parties) and to transfer their preferences to opposition and new parties. The total volatility will then increase.
6 Table 1. Total electoral volatility pre-crisis and during crisis (to be completed with CEE countries) Country Increase (%) Austria 11, Belgium 11,29 12,82 13,5 Denmark 9,92 15,2 53,2 Finland 10 11,47 14,7 France 20,03 23,6 17,8 Germany 8,1 15,22 87,9 Greece 6,15 24,41 296,9 Ireland 8,85 27,2 207,3 Italy 13,62 23,97 75,9 Netherlands 21,24 19,7-7,2 Portugal 11,36 11,13-2,0 Spain 8,47 12,13 43,2 Sweden 15,68 9,77-37,7 UK 8,05 12,7 57,7 Average volatility 11,7 16,88 44,2 Source: V. Emanuele, 2015 and personal elaboration This analysis which does not include at this stage central eastern European countries indicates an increase in volatility in 11 of 14 countries (Table 1). Greece and Ireland with volatility tripling from the first to the second period are the two most striking cases followed by Germany and Italy, then by the UK, Austria, Denmark and Spain. Sweden and to a lesser extent Netherlands and Portugal, show on the contrary a decreased
7 volatility. The quantitative analysis conducted with measures of electoral volatility could be improved by more qualitative instruments, as by exploring to what extent the increased volatility is due to the rise of new parties or of parties that are critical of the establishment and/or of Europe, instead of shifts between established parties. While volatility has increased more or less everywhere during the second period a simple inspection shows that for western EU countries the relationship between increase in volatility and depth of the crisis is not equally strong. Just to mention a few cases, among the countries most heavily hit by the crisis Portugal shows low levels of volatility. On the other side Germany, one on the countries which has fared best in the crisis has seen a rather significant increase in volatility. The suggestion that can be derived from this is that the effect of the crisis is probably mediated by other variables which may reduce or increase its impact. The resilience of the party system is probably one of these. More research along these lines is required. If increases in volatility signal a greater readiness of voters to shift allegiances, the next step is to ask to what extent this means also a greater readiness to punish governments, presumably for their inability to solve the problems caused by the crisis(es) or at least to compensate their effects. To investigate this aspect I have used two different indicators : a. the electoral losses of parties represented in government (Parties in Government Electoral Losses or PiGEL); b. the change in the person of the Prime Minister (PMChan) after an election.
8 Table 2 Indicators of Cabinet vulnerability Country PiGEL PiGEL PMCHAN PMCHAN % % Austria 2,7 8,6 3/4 1/2 Belgium 0,4 2,2 2/4 2/2 Bulgaria 26,5 12,3 3/3 3/3 Czech Republic 1,5 21,53 3/4 2/2 Croatia 9,5 12,55 3/4 2/2 Denmark 2,1 4,25 1/4 2/2 Estonia 0,95-1,2 3/4 0/2 Finland 3,9 7,7 2/4 2/2 France -1,6 12,6 1/3 0/1 Germany 3,8 7,3 2/3 0/2 Greece 2,5 21,7 1/4 4/4 Hungary 6,7 17,8 2/3 1/2 Ireland 1,8 26,5 1/3 1/2 Italy 0 4,3 1 3/3 2/2 Latvia 8,4 6,9 3/4 0/3 Lithuania 16,4 9,8 2/3 2/2 Netherlands 9,3 12,6 1/4 1/2 Poland 19 10,6 4/4 1/2 Portugal 8,9 9,6 3/4 2/3 Romania 10 8,4 3/3 2/2 Slovakia 14,5 12,7 2/3 2/3 1 In the Italian case I have counted as governing parties also the parties which gave parliamentary support to the Monti technical government.
9 Slovenia 11 19,2 1/3 3/3 Spain 0,6 16,1 2 2/3 1/2 Sweden 3,4 4,2 1/3 1/2 UK 6,4 10,2 1/3 1/2 Average 6,7 11,1 61% 68% PiGEL: a positive % indicates an average electoral loss; a negative % indicates an average gain PMChan: the numbers indicate the proportion of new prime ministers over all the governments formed after an election The data collected for all the 28 EU countries show in a large majority of cases for the PiGEL indicator the negative impact of this period over the electoral performances of parties in government. In 21 out of 28 cases the parties in government have on average suffered more during the second period than during the first one. And the average losses of governing parties in all countries have nearly doubled between the first and the second period. It is interesting to notice that in 7 of the 11 CEE and Balkan members of the EU the losses of the parties in government were in the second period smaller than in the first one. This result, which might appear surprising, is probably to be explained by the fact that in the new democracies the first period involved the difficult problems of the transition and a lengthy process of institutionalisation of the pluralistic party system: these two factors have probably affected negatively the performances of parties in government. Among the old Europe states three of the countries most heavily hit by the economic crisis - Greece, Ireland and Spain - have seen the greatest 2 The elections of 2016 were not counted as the cabinet in charge before the elections was a caretaker government
10 average losses of the governing parties. It is an open question why the impact has been much lower in Italy and Portugal, which have been also deeply affected by the crisis. With regard to the second indicator the change of Prime minister between the cabinet preceding the elections and the one following them (PMCHAN) the table indicates that in more than 60% of the cases the incumbent Prime ministers do not survive an election. In this case also the average has (moderately) increased between the first and the second period. The first indicators used to assess the vulnerability of national elites of the EU during this period have consistently shown a negative change between the first and the second period under exam. The volatility of electorates has increased almost everywhere. The governing parties and the Prime ministers have also faced greater difficulties in defending their positions. The European level effects Given the institutional nature of the European Union the European level effects are to a large (but not full) extent the consequences of national level effects. If we consider the European council or the Council of ministers being its composition indirect it will be obviously affected by all the changes in national governments. In order to assess how much these bodies are affected by national changes we can look at the changing composition of their membership due to changes in government taking place in the member states.
11 For the European Council I have computed an Index of European Council turnover. As the number of members of the Council has changed over time due to accessions of new countries the Index is calculated as the yearly percentage of new EC members divided by the total number of members) 3. Between 1995 and 2007 the index was on average of 17.3%; from 2008 to 2016 it increased to 22.6%. With this turnover rate in two years the membership of the European Council is renewed by nearly a half. We can then say that if the European Council, as well as the Council of the European Union, are by their nature institutions with a constantly changing membership, an increased vulnerability of national elites and national governments makes for a more rapid rotation of their membership. Although this is not the only possible explanatory factor for their institutional performance, should we be surprised if the ability to decide promptly of the European Council when faced with crisis situations is reduced? How can frequently changing compositions of that body enable the reaching of agreements? What has happened to other components of the EES during this period? This part of the analysis is left to a new version of this paper. We can anticipate just one point. The executive board of the ECB, whose members are required not to represent their country of origin, but who probably share some cultural and policy orientations typical of their motherland, has emerged as one of the most stable and cohesive elite groups in the EU. But even this group has paid some price to the special challenges and pressures of this period. This is suggested by a few cases of anticipated resignations; in particular are to 3 As in the year of accession of a new country its member of the Council is necessarily new I have started counting it from the second year of participation. For example the member of the European Council for Austria (which entered in the EU in 1995) was counted only in 1996.
12 be mentioned the two resignations in a row of the German member (first Stark then Asmussen). Yet the central decision making body of the ECB is to be considered a very stable and durable component of the EES bringing to it a strong technocratic component but not devoid of remarkable policy vision. Preliminary conclusions I will refrain here from sweeping conclusions as the research for this paper is still under way. In any case the data presented so far give some preliminary indication about the effects of the crisis over the complex European Elite System and about the paradoxes of its structure. On one side the decisional centre of gravity is kept (or even increased) in the intergovernmental institutions who are precisely those most negatively affected by the crisis and the least able to produce efficient decisions. Thus a deficit of democratic decisions. On the other side there are complaints about the role of technocratic institutions as non-democratic, but these institutions are de facto pushed to substitute for the more democratic institutions if these are not able to decide. Should we be surprised if this situation strengthens the general dissatisfaction with the EU? References Anderson, C. D. (2006). Economic voting and multilevel governance: a comparative individual level analysis. American Journal of Political Science, 50 (2),
13 Bellucci, P., & Lewis Beck, M. S. (2011). A stable popularity function? Cross national analysis. European Journal of Political Research, 50(2), Cotta, M. (2012) Political Elites and a Polity in the Making: The Case of the EU. In Historical Social Research, vol. 37, pp Cotta, M. (2014) Facing the Crisis: The European Elite System s Changing geometry, in H. Best and J. Higley eds., Political Elites in the Transatlantic Crisis, Houndmills, Palgrave, pp Emanuele, V. (2015), Dataset of Electoral Volatility and its internal components in Western Europe ( ), Rome: Italian Center for Electoral Studies, Fabbrini, S. (2007), Compound democracy, Oxford, Oxford University Press. Hooghe, Lisbet and Gary Marks 2001, Multi-level governance and European integration, Lanham, Rowman and Littlefield Publishers. Lago, I., & Martínez, F. (2011). Why new parties?. Party Politics, 17(1), Lewis Beck, M. S., Nadeau, R., & Elias, A. (2008). Economics, party, and the vote: Causality issues and panel data. American Journal of Political Science, 52(1), Lewis-Beck, M. S., & Nadeau, R. (2011). Economic voting theory: Testing new dimensions. Electoral Studies, 30(2), Oesch, Daniel. "Explaining Workers' Support for Right-Wing Populist Parties in Western Europe: Evidence from Austria, Belgium, France, Norway, and Switzerland." International Political Science Review 29, no. 3 (2008): Tavits, M. (2006). Party system change testing a model of new party entry. Party Politics, 12(1), Tavits, M. (2008). Party systems in the making: The emergence and success of new parties in new democracies. British Journal of Political Science, 38(01),