In practice this indicator comes down to the role of political parties in intermediating between citizens and the political system. Political parties are needed to structure voter choice in elections. In an ideal world voters would have just that choice in elections that allows them to choose in the way in which they want to choose, whether that is a matter of choosing between alternative policies or choosing between alternative leaders. However, even in perfectly competitive party systems, it would be difficult for parties to anticipate exactly how voters want to make their choices; and in practice; of course, most party systems are less than perfectly competitive. Even allowing for the recent success of protest parties, there are significant barriers in many systems to the entry of new political parties to the process of electoral competition or to the reconfiguring or realignment of existing parties. This means that many voters are constrained in their choices by the range of parties, policies and leaders on offer.
In the case of the European Union there is a further complicating factor: in effect the Union depends on a split-level party system. That is to say national parties structure choice even for European Union. Yet the work of the European Parliament itself is structured by European party groups. Table 3.4 shows which national parties belong to which European party groups. For the moment, though, it is enough to note that if we are to evaluate how parties structure voter choice and then translate those choices into the workings of Union institutions we will need data on both national parties and the European party groups.
Amongst measures which could be relevant for measuring choice and political competition are the following: proximity to political parties (3.1); trust in political parties (3.2); inclusiveness of the representation of national political parties in the European Parliament (3.3); fit between the national and European party systems (3.4); and cohesion of party groups in the European Parliament (3.5).
3.1 Proximity to political parties
One measure of whether voters feel they have a satisfactory choice is provided by a question that has long been asked by opinion polls: namely, whether they feel close to a political party. This is a problematic question, since different people may answer it in different ways. Some may interpret the question as asking them whether they identify with a political party. Others may see it as asking them whether there is a political party that has policies they can agree with. Still, it does provide us with some measure of whether voters feel that there is at least one party on offer for which they can vote.
Across the Union as a whole, the number of citizens who do not feel close to any one political party outnumber those who do. However, variation across member states is huge, as shown in the last column of Table 3.1.
3.2 Trust in political parties
It is not, however, enough for voters to feel close to at least one political party. They also need to trust political parties to deliver. Table 3.2. shows that there is not a single EU member state in which more citizens trust political parties than do not. The average of those who trust political parties across the Union as a whole is just 15 per cent. Mistrust of political parties is high in several post-communist member states where parties are often seen for historical reasons as instruments of domination, rather than mechanisms for voter choice.
However, it would be a mistake to assume that mistrust of political parties is confined only to post-communist states. A comparison of Table 3.2. with Table 3.1. reveals another pattern: trust in political parties and feelings of proximity to them are by no means always related. In some member states political parties are relatively less mistrusted. Yet voters don’t feel especially close to them. Denmark is an example. In other member states, voters feel that close to at least one political party. Yet they mistrust political parties in general. Italy is an example of this.
3.3 Inclusiveness of the representation of national political parties in the European Parliament
RECON Model 1 would presumably imply that the powers of the European Parliament are best understood as a delegation of powers from national parliaments. That it has, in other words, been delegated with the powers to monitor other EU institutions on behalf of national parliaments. Maybe, for example, national parliaments would face substantial asymmetries of information in seeking to control Union decisions that can be best overcome by delegating the task to a dedicated European Parliament that can specialise in following Union decisions and building up the expertise needed to do that.
If, however, the EP is best justified as ‘standing in for national parliaments’ – as a body that monitors Union decisions on behalf of national parliaments – then it would presumably be helpful if it also represented a similar range of national parties to those found in national parliaments.
The third column of Table 3.3. shows that it does, indeed, include a wide range of the national parties found in most member states. The first and second column of the table compares the effective number of parties represented in national parliaments and the EP. In other words, it also allows for variation in the size of the representation given to different national parties at the two levels. Here member states vary. In some cases, the effective number of parties represented in the EP is higher than in national parliaments. In others it is lower.
3.4 Fit between the national and European party systems
As seen, the EU operates a two-level party system. National parties structure voter choice in European elections. Yet European party groups exercise the powers of the European Parliament once it is elected. That raises the question of the ‘goodness of fit’ between party systems at the two levels. One possible measure is provided by how often national party delegations are able to vote the same way as their European party groups.
VoteWatch provides an invaluable resource to researchers in compiling all the roll-call votes of the European Parliament in an easily searchable form. Table 3.4. shows each European Party group in the 2009-14 Parliament and each of its component national parties. The table shows how often the majority of each national party delegation voted with the majority of its European Party Group in the roll-calls taken during the first two years of the 2009-14 Parliament. Only those national party delegations highlighted in blue failed to vote with their party groups in 90 per cent of roll-calls or more. Only 67 MEPs – or less than 10 per cent of the Parliament – belong to those national party delegations, and 36 of those 67 MEPs are concentrated on just two groups: the far left group and the Eurosceptic group. For the most part, then, there would appear to be a fairly good fit between the national party delegations and the European party groups.
3.5 Cohesion of party groups in the European Parliament
It is a matter of some dispute how far cohesive and disciplined political parties are desirable from a point of view of democratic standards. Party disciplines can stifle political debate and constrain individual representatives in the representation of their own voters.
Yet from at least one point of view party disciplines can be helpful. Individual representatives come and go. Yet political parties often continue for a long time. This may mean that political parties can be held accountable over long periods of time in ways in which individual representatives cannot. On top of that, parties compete around broad programmes of government. Thus they can be held accountable for how they aggregate individual issues. This provides voters with some choice control over how value trade-offs are made across the range of public policies.
For these reasons, many have argued that representative democracy is really a system of ‘party responsible government’. Parties need to be able to control individual representatives if parties as a whole are to be responsible to the voters. At least from this point of view it may be desirable that representatives are able to form coherent party groupings in the European Parliament. Table 3.5. uses roll-call data to show cohesive the EP party groups are.