Valence Advantages and Public Goods Consumption: Does a Disadvantaged Candidate Choose an Extremist Position?

Does a disadvantaged candidate always choose an extremist program? When does a less competent candidate have an incentive to move to extreme positions in order to differentiate himself from the more competent candidate? If the answer to these questions were positive, as suggested in recent work (Ansolabehere and Snyder (2000), Aragones and Palfrey (2002), Groseclose (1999), and Aragones and Palfrey (2003)), this would mean that extremist candidates are bad politicians. We consider a two candidates electoral competition over public consumption, with a two dimensional policy space and two dimensions of candidates heterogeneity. In this setting, we show that the conclusion depends on candidates relative competences over the two public goods and distinguish between two types of advantages (an absolute advantage and comparative advantage in providing the two public goods).