If there is no evidence of a religious resurgence in Turkey, what can explain the rise and sustained success of Islamic-based parties there? To understand the popularity of Islamic parties, like the AKP, a broader view of the Turkish electoral system as warranted: alongside the rise of the AKP came a sharp decline in electoral volatility -- vote swings between parties from election to election -- and in the share of votes that were wasted, cast for parties that failed to secure a seat in a given district. I argue that these two trends are not coincidental but are both based on matters of trust: low levels of interpersonal trust makes it difficult for voters within districts to vote strategically and successfully coordinate their individual votes into meaningful outcomes; but this trust problem is effectively solved within religious voters, to the comparative advantage of Islamic parties. Moreover, the ability of religious voters to coordinate their support for Islamic parties, and to do so consistently, helps to make these parties an attractive target for strategic votes from distrusting, conservative voters, even if they are secular. Analysis of panel data from the Turkish case provides empirical support for both hypotheses.

We examine the notion of linguistic and other types of diversity that have become an important factor in evaluating economic, political, and societal progress. While most of the existing research on the measurement of diversity has been focused either on the number and size of different groups, this approach fails to take into account the degree of their distinctiveness. It is thus important to incorporate the notion of distances or dissimilarity between groups. We discuss various types of indices and their impact (positive or negative) on economic and political outcomes. We also use linguistic distances to explain economic behavior such as trade, migrations, translations, and acquiring new languages.