The opinion formation and the decision of the citizens
Given the key role of the political elites in the direct-democratic process in Switzerland, we already have a partial answer to the question of how the citizens are able to make reasonable choices at the polls, even if they are politically relatively uninformed. The political elites define the options, which they submit to the citizens’ vote; they form the coalitions that oppose each other in the campaign preceding the vote; and the mobilization of these coalitions supplies the heuristic cues and the systematic arguments that the citizens use for opinion formation during the campaign. The remaining question is whether the citizens are capable of properly using the information supplied to them by the elites, and whether they are able to arrive at a choice that reflects their own preferences.
According to my own study of roughly 150 popular votes, I have come to rather optimistic conclusions in this regard. First of all, the share of the electorate that is moderately to fully aware of the proposals submitted is rather high. The overall average for the 217 federal propositions of the period 1981-2004 reaches a level of no less than 73 per cent. It is true that there are exceptions to this high level of awareness. In almost 10 per cent of the cases, a majority of the electorate was quite unaware of the issues in question. Moreover, the level of awareness has been decreasing over the years, from an average of 80 per cent in the 1980s to an average of 69 per cent in the 2000s. However, citizens who actually participate in the vote prove to be significantly more competent than the electorate overall. This means that the most incompetent usually do not participate in the vote. This mechanism of self-selection of the incompetent reduces in a quasi-automatic way the possibility of an unreasonable decision, which the critics of direct-democratic procedures are so much afraid of. Even if it is true that more intense campaigns mobilize a larger number of voters from all camps, that is, that they also mobilize voters who are usually rather uninformed, the participation of those who are little interested and uninformed remains limited even in such campaigns. This results from a self-correcting mechanism whereby particularly intense campaigns that mobilize large numbers of citizens also serve to raise the general level of issue- specific competence in the electorate at large - in turn, implying that, in such cases, the group of uninformed voters becomes relatively small.
Those members of the electorate who actually participate in the vote have - following a well known theory of social psychology (the so called ‘dual process theory’, Stroebe 2007) - basically two possibilities for arriving at their choice: either they rely on heuristic cues such as voting recommendations by parties, trust in the government or defence of the status quo, which allow them, in a short-cut way, to arrive at approximately reasonable decisions. Or they may more systematically rely on the arguments promoted by the adversarial camps. In the real world of opinion formation and decision making in direct-democratic campaigns, however, the two analytically distinct ways of deciding are not so sharply distinct. On the one hand, the effective use of heuristic cues presupposes a certain amount of political knowledge. Thus, one has to know about the general positioning of the political parties in order to be able to make effective use of their recommendations. On the other hand, some of the arguments that are used by the political elites in the course of the campaigns are hardly distinguishable from heuristic cues. Thus, the decision of voters who rely on the often all too simplistic arguments that are promoted in the campaign, will not be much different from a purely heuristically determined decision.
Even if the two paths are not always as clearly distinct as the theory would have it, the results of my study with regard to the citizens’ capacity to arrive at reasonable decisions are nevertheless quite clear-cut. The citizens turn out to
Figure 2.5 Probability of governmental support, as a function of argument-based opinions Source: Kriesi 2005 be less minimalist than usually assumed: my study confirms the generally great importance of argument-based decisions. It measures argument-based opinions on the basis of a set of questions about the most important arguments exchanged during the campaigns preceding the vote. A voter’s overall argument-based opinion score corresponds to his or her positioning on the central conflict dimension of the campaign-specific debate (which is operationalized by the first factor resulting from a factor analysis of the opinions on the campaign-specific set of arguments). The great importance of argument-based voting is illustrated by Figure 2.5, which shows the general relationship between argument-based opinions and the voting choice, that is, the support of the government’s position: the more strongly citizens support the arguments in favour of the government’s position, the more strongly they also decide in favour of the government’s position at the polls - and vice versa. The relationship is rather tight, which means that we can explain roughly 40 per cent of the variance of individual voting decisions on the basis of the voters’ opinions about the arguments that were prevalent during the campaign.
The importance of arguments varies, however, from one campaign to another depending on its intensity and on the familiarity of the proposal. Compared with these two context charact eristics, individual characteristics play a subordinate role for the explanation of argument-based voting. The second half of Figure 2.5 presents the maximum joint effect of the two context characteristics on the relationship. As we can see, the curve is much flatter when both the intensity of the campaign is low and the proposal is unfamiliar. This means that arguments turn out to be much more decisive in highly intense campaigns and for highly familiar propositions - especially, as the figure also indicates, as far as the challengers of the government’s position are concerned. They benefit from intensive campaigns more than the government. While the campaigns are reasonably intense in the large majority of cases, a majority of cases also proves to be rather unfamiliar to significant minorities of the electorate (a third or more). Overall, roughly a quarter of the propositions was both rather unfamiliar and did not give rise to intensive campaigns. These were also the cases where the level of awareness turned out to be particularly low, with an average share of the electorate of only 59 per cent being at least moderately aware.
Systematic and heuristic strategies are to a certain extent complementary. Thus, intensive campaigns increase the relevance of both decision strategies. The two strategies are also complementary in another respect: they are used in the same context, but by different types of voters. While the voters with strong opinions - either on the side of the opposition or on the side of the government - generally decide systematically on the basis of arguments, voters with less explicit opinions, that is, ambivalent, uncertain, neutral or ignorant voters, more heavily rely on heuristic strategies. This is illustrated by Figure 2.6, which is divided into three parts. Left and right from the centre, one finds the voters with strong opinions; in the centre, those with less explicit opinions. In each part of the figure, the relationship between the level of awareness and the support of the government’s position is shown for voters with different partisan orientations and different levels of trust in the government. Without going into the details of the figure, one can easily recognize that the lines for voters with strong opinions on both sides of the centre can hardly be separated from one another, which means that they all decide in roughly the same way, independently of their partisan orientation or their level of trust in the government. For the voters with less explicit opinions, by contrast, the lines for the different groups of voters are clearly distinguishable from one another, meaning that these voters more heavily rely on partisan recommendations and their trust in government - two types of heuristic cues - to arrive at their voting decisions. Since the figure refers to propositions that are supported by centre-right coalitions and opposed by the left, the trusting conservative voters support the government most, while the distrusting left voters do so least. Finally, the figure also clearly shows that, for virtually all types of voters, the lines more or less run parallel to the x-axis. This means that the level of awareness has practically no impact on the voting behaviour. Even little-informed voters are, in other words, capable of arriving at decisions that closely resemble those of their well-informed colleagues with similar preferences (that is, party orientations and levels of trust in government).