The most famous result in political science (Downs 1957) is that in a two candidate electoral race the two contestants should move toward the middle of the political spectrum. More precisely they should propose the policy preferred by the median voter. Given that this is the best strategy for the two candidates, they should be as clear as possible on what they are doing and eliminate every uncertainty about their platforms.
Nothing seems further from real world elections. In many two party elections, candidates are very polarised, often leaving a large ideological hole in the middle. The current American election is an example: the two candidates disagree on foreign policy, abortion, health care, taxation, you name it. The previous two presidential elections were even more polarised. In recent French, Italian, and Spanish elections, the two blocs of parties presented distinctly different platforms.
The prediction that electoral platforms should be clear and unambiguous is also in sharp contrast with reality. Politicians are famous for their verbal contortions in order to avoid taking a clear stand in the run up to the election. The ambiguity takes two forms. One is simply being vague; the second is adjusting the message depending on the audience.
How do we reconcile the prediction of the canonical model of two party elections with reality? In a recent paper (Alesina and Holden 2008), we argue that there are two forces pulling parties in opposite directions. The standard force pulls the parties toward the middle – conquering the moderate vote given that the left (right) candidate has a solid hold of the left (right) wing vote. But there is another force that pulls parties apart. And this is the effect of campaign contributions, broadly defined in the sense of money, time of activists, strikes of friendly unions etc.
Campaign contributors (in that broad sense) tend to be rather extreme, namely away from the middle of the political spectrum. Their contributions may move the mood of the electorate toward their political side. For instance money coming from conservative groups can buy TV time that moves middle of the road voters toward the right. Time spent by activists of the left can make the left wing party gain votes in critical districts. Thus, for example, a right wing candidate has to balance these two forces. By moving to the right, he or she can garner more conservative votes and move the median voter to the right. By moving to the middle, he or she loses contributions but moves closer to the median voter who may be less on the right. The balance of these two forces leads to a choice that is not to run to the middle. The same reasoning applies to a left-leaning candidate, and so you have a polarised equilibrium.
The same principle implies ambiguous platforms. In a sense, the candidates would like to have their cake and eat it too. Rather than announcing a precise platform, they may announce a range of policies. The contributors would hope that once elected the candidate would choose the closest to them (i.e. the most extreme of the range). The middle of the road voters would hope the opposite. With this ambiguous balancing act, the candidate can be better off by garnering more contributions and losing fewer middle-of-the-road voters relative to taking a precise policy stance. Contributors and voters can be aware of the candidates’ incentives, but as long as the candidates can hide their true preferences, in equilibrium (even with full rationality and risk aversion of voters and contributors) it can be preferable for the candidates to remain ambiguous.
These considerations apply to any two-party election. In the American system, primaries are an additional force that leads to ambiguity in the process. Each candidate would like to keep an option value of his or her policy stance. By remaining ambiguous in the primaries, politicians can then adjust their positions as a function of who their opponent will be. Thus, uncertainty concerning the opponent’s identity and policy stance induces further ambiguity. Given both parties’ incentive to be ambiguous, learning the name of your opponent in general would not fully reveal his or her policy position. Therefore adopting a clear and unambiguous position in the primaries may be too much of a straight jacket in the presidential election. On the other hand risk-averse voters would not like too much ambiguity in the primaries. The balance of these two forces leads to a certain amount of ambiguity in primaries to be added to the incentive to be ambiguous in all two party contests.
Ambiguity during hard-fought primaries can reach almost comical levels. For instance in the latest Republican primaries, John McCain accused his rival Mitt Romney of changing his position on abortion every even-numbered year. His answer was that his views on abortion changed because of new discoveries in the research on cloning! (sic)
Alesina and R. Holden (2008) “Ambiguity and Extremism in Elections”, NBER Working Paper
Downs, Anthony (1957). An Economic Theory of Democracy, Harper and Row, New York, NY.