The Electoral System

This Latin American sample offers a different set of cultures, histories, and social cleavages. It suggests that the effect of electoral system choice on multipartism translates significantly across the limitation of social factors. Even though Jones’s data on Latin American countries shows a substantially higher level of disproportionality than Lijphart’s data, the demonstration of electoral systems’ influence on party system variables is shared across both sets. Based on their data, the electoral system variables of formula and constituency magnitude appear to be electoral systems’ most powerful shaping forces. Though Morgenstern and Vázquez-D’Elía (2007) did not corroborate this correlation, they studied only PR systems, and the absence

When one says that electoral systems shape the party system, it must be said with the restraint of understanding that electoral systems do not exist unless selected by partisan entities competing in elections. The opportunity for electoral systems to shape party systems thus arises because of parties’ decisions. If the party system variable of multipartism influences which electoral system is selected, then this raises problems studying the extent to which electoral system variables like disproportionality affect multipartism—this being the endogeneity problem (Benoit 2007). This does not mean that electoral systems do not shape party systems merely because parties select electoral systems. What it does mean is that the extent to which party systems shape electoral systems in return may be relevant to how much influence electoral systems can exert. Benoit (2007) takes this line of reasoning to its important logical conclusion: If parties have a causal role because they select the electoral system, does this not validate the argument that electoral systems shape party systems? For what reason would parties change the electoral system if not to manipulate the electoral system variables in a meaningful way that benefits their partisan interests? Again, it seems that this

Data on Western and Latin American democracies gives statistical support to relationship, and exceptions like the UK are convincingly explained by external social factors. Further, possible arguments against the very idea that it is possible for electoral systems to shape party systems only appear to show that causality in the framework is confused, not that the measured impacts of electoral systems are not