Kantian theories of morality focus on the universal application of moral rules. However, both children and adults often apply different moral standards to in-group and out-group members. Psychologists have proposed that this group bias in moral judgments may be explained by “social essentialism”, a tendency to conceive of social groups as natural kinds. This dissertation uses a cross-cultural, developmental approach to test this hypothesis by investigating a) how the essentialization of five social groups changes with age and b) whether the differences in essentialization explain children’s moral judgments in inter-group contexts. In Study 1, I tested the degree of essentialization of five social categories (Gender, Nationality, Religion, Socioeconomic Status (SES), and Teams) in 5-10 year olds (N=147) and adults (N=223) in Turkey and the U.S. I hypothesized three possible patterns of results indicating different mechanisms underlying essentialization: 1) essentialization is a strong basic bias invariant across ages, cultures and categories; 2) essentialization varies by category across culture based on historical group conflicts; and 3) essentialization is over-generalized for pseudo-biological categories (Gender, Nationality) and declines with age for other categories in both cultures. I found strong support for the third mechanism and striking similarities in the developmental patterns by category across cultures. Study 2 examined the hypothesized link between children’s social essentialist bias and moral judgments in the US (N=211). I predicted that for highly essentialized categories from Study 1 (i.e., Gender), children would believe that it is more acceptable to harm the out-group than the in-group. There were no systematic differences between in-group and out-group judgments and no relationship with essentialization, however. These null results suggest that children are more Kantian than recent work on social groups proposes. Essentialism did, however, affect moral reasoning in inter-group contexts in more indirect ways, when accompanied by other social phenomena, such as salient discrimination. Combined, these studies make two contributions to the field. First, essentialist beliefs in the social domain are triggered cross-culturally by a biological representation of some categories. Second, children are not generally sensitive to group membership in their explicit judgments of moral transgressions in third-party scenarios.