This paper presents an economic model of friendship formation that explains segregation patterns in social and economic networks. The model considers how individuals form friendships based on their types and the benefits they derive from such relationships. The authors examine three empirical patterns of friendship formation: (1) larger groups form more same-type ties and fewer other-type ties than small groups, (2) larger groups form more ties per capita, and (3) all groups are biased towards same-type friendships relative to demographics, with the most extreme bias coming from middle-sized groups. The model shows how these observations can be generated by biases in preferences and biases in meetings. It also illustrates some welfare implications of the model. The paper discusses homophily, which refers to the tendency of individuals to associate with others who are similar to themselves. Homophily has been documented across various characteristics such as age, race, gender, religion, and profession. The authors use the model to analyze homophily in the "Adolescent Health" data set, which examines friendship patterns in a representative sample of U.S. high schools. They identify three main observations: (1) larger groups form a greater fraction of same-type friendships, (2) larger groups form more friendships per capita, and (3) groups tend to form same-type friendships at rates that exceed the relative fractions in the population, with the most extreme bias coming from middle-sized groups. The authors develop a model of friendship formation where individuals form friendships when they enter a society and meet through random matching. Each matching entails a fixed cost, and there are diminishing returns to forming friendships. The critical determinants of an individual's strategy are their preferences over the mix of types of friends they form and the mix they face in the matching process. The model shows that if agents' preferences over friendships are insensitive to type, then all agents form the same number of friendships under any matching process. However, if agents have type-sensitive preferences, then larger groups form more friendships per capita in equilibrium. The authors also show that a bias in the matching process, so that all groups can simultaneously be meeting their own type at rates faster than their fraction in the matching pool, generates inbreeding homophily patterns that match the observed patterns. Such a bias in matching could be attributed to factors such as tracking and membership in various clubs and activities, as well as meeting friends through friends. The authors fit a parametric version of the model to the data and show that a model with both type-sensitive preferences and a matching bias fits all of the observations, while just type-sensitive preferences generates the observation of larger groups forming more per capita friendships but not the inbreeding homophily, while just a matching bias does the reverse. The paper also discusses welfare implications of the model, noting that while the model is too stark to take seriously for policy implications, the welfare analysis points out that average welfare depends in sensitive ways on the structure of preferences and matching. This suggests that it is vital toThis paper presents an economic model of friendship formation that explains segregation patterns in social and economic networks. The model considers how individuals form friendships based on their types and the benefits they derive from such relationships. The authors examine three empirical patterns of friendship formation: (1) larger groups form more same-type ties and fewer other-type ties than small groups, (2) larger groups form more ties per capita, and (3) all groups are biased towards same-type friendships relative to demographics, with the most extreme bias coming from middle-sized groups. The model shows how these observations can be generated by biases in preferences and biases in meetings. It also illustrates some welfare implications of the model. The paper discusses homophily, which refers to the tendency of individuals to associate with others who are similar to themselves. Homophily has been documented across various characteristics such as age, race, gender, religion, and profession. The authors use the model to analyze homophily in the "Adolescent Health" data set, which examines friendship patterns in a representative sample of U.S. high schools. They identify three main observations: (1) larger groups form a greater fraction of same-type friendships, (2) larger groups form more friendships per capita, and (3) groups tend to form same-type friendships at rates that exceed the relative fractions in the population, with the most extreme bias coming from middle-sized groups. The authors develop a model of friendship formation where individuals form friendships when they enter a society and meet through random matching. Each matching entails a fixed cost, and there are diminishing returns to forming friendships. The critical determinants of an individual's strategy are their preferences over the mix of types of friends they form and the mix they face in the matching process. The model shows that if agents' preferences over friendships are insensitive to type, then all agents form the same number of friendships under any matching process. However, if agents have type-sensitive preferences, then larger groups form more friendships per capita in equilibrium. The authors also show that a bias in the matching process, so that all groups can simultaneously be meeting their own type at rates faster than their fraction in the matching pool, generates inbreeding homophily patterns that match the observed patterns. Such a bias in matching could be attributed to factors such as tracking and membership in various clubs and activities, as well as meeting friends through friends. The authors fit a parametric version of the model to the data and show that a model with both type-sensitive preferences and a matching bias fits all of the observations, while just type-sensitive preferences generates the observation of larger groups forming more per capita friendships but not the inbreeding homophily, while just a matching bias does the reverse. The paper also discusses welfare implications of the model, noting that while the model is too stark to take seriously for policy implications, the welfare analysis points out that average welfare depends in sensitive ways on the structure of preferences and matching. This suggests that it is vital to