Politics has always attracted the attention of the media, citizens organizations, and the general public. Recent years have also witnessed a global process of “spectacularization” of politics, which, among other things, has resulted in a dramatic increase in the amount of information available about many facets of political life. Politicians, for example, are public figures, and much of what they do is now the object of close public scrutiny. Nevertheless, the extent to which various aspects of what goes on within the political sector are observable from the outside, which we refer to as the transparency of politics, still varies a great deal across countries. For example, while in some countries all individual votes in the legislature are part of the public record (e.g., the United States and Sweden), this is not the case in others (e.g., Italy and Spain). Also, while many democracies have adopted disclosure laws that require political parties and politicians to report all the contributions they receive (e.g., Canada and the United Kingdom), such laws are not in place in several other countries (e.g., Austria and Finland). It is therefore interesting to ask whether the transparency of politics may be systematically related to political outcomes, and whether more transparency would lead to better outcomes. In particular, in this article, we analyze the relationship between the transparency of politics and the quality of politicians, and focus on the recruitment of politicians by political parties. Parties represent a fundamental institution of representative democracy, and are the political-sector analogues of firms in the market sector. By and large, politicians are affiliated with a party, and typically start their political careers by working for party organizations (see, e.g., Heinrich Best and Maurizio Cotta 2000). Hence, the recruiting decisions of parties determine the quality of the pool of politicians.

We introduce a model where social norms of voting participation are strategically chosen by competing political parties and determine voters’ turnout. Social norms must be enforced through costly peer monitoring and punishment. When the cost of enforcement of social norms is low, the larger party is always advantaged. Otherwise, in the spirit of Olson (1965), the smaller party may be advantaged. Our model shares features of the ethical voter model and it delivers novel and empirically relevant comparative statics results. (JEL D72, Z13)