This paper presents an alternative model of agglomeration and trade that captures the main features of recent economic geography literature while allowing for analytical results. It shows how this framework can be used to adopt a forward-looking approach to the dynamics of migration in the process of agglomeration, rather than the myopic Marshallian model used in the literature. The model is based on a quadratic utility function that allows for product differentiation in imperfect markets. It differs from existing literature in several respects, including the use of a different utility function, closed analytical solutions, and the ability to model asymmetry between own price and cross price effects. The model also allows for a more detailed analysis of the forces at work and more explicit connections to industrial organization literature. The paper shows that expectations influence the agglomeration process in a totally unsuspected way, as they have an influence on the emergence of a particular agglomeration for intermediate values of the transport cost only. For such values, if workers expect the lagging region to become the leading one, their expectations will reverse the dynamics of the economy provided that the difference in initial endowments between the two regions is not too large. The paper also shows that the model allows for a more detailed analysis of the forces at work and more explicit connections to the industrial organization literature. The model is more natural than using the iceberg transport cost and is in accordance with what has been accomplished in spatial pricing theory. The equilibrium concept used is broader than that employed by Dixit and Stiglitz in that it agrees with Chamberlin while yielding prices depending on all the fundamentals of the market. This permits a more detailed analysis of the forces at work as well as more explicit connections to the industrial organization literature. The paper concludes that the model may be considered as an alternative to the 'Dixit-Stiglitz, iceberg, evolution, and the computer' framework which has served as the main tool-box in economic geography.This paper presents an alternative model of agglomeration and trade that captures the main features of recent economic geography literature while allowing for analytical results. It shows how this framework can be used to adopt a forward-looking approach to the dynamics of migration in the process of agglomeration, rather than the myopic Marshallian model used in the literature. The model is based on a quadratic utility function that allows for product differentiation in imperfect markets. It differs from existing literature in several respects, including the use of a different utility function, closed analytical solutions, and the ability to model asymmetry between own price and cross price effects. The model also allows for a more detailed analysis of the forces at work and more explicit connections to industrial organization literature. The paper shows that expectations influence the agglomeration process in a totally unsuspected way, as they have an influence on the emergence of a particular agglomeration for intermediate values of the transport cost only. For such values, if workers expect the lagging region to become the leading one, their expectations will reverse the dynamics of the economy provided that the difference in initial endowments between the two regions is not too large. The paper also shows that the model allows for a more detailed analysis of the forces at work and more explicit connections to the industrial organization literature. The model is more natural than using the iceberg transport cost and is in accordance with what has been accomplished in spatial pricing theory. The equilibrium concept used is broader than that employed by Dixit and Stiglitz in that it agrees with Chamberlin while yielding prices depending on all the fundamentals of the market. This permits a more detailed analysis of the forces at work as well as more explicit connections to the industrial organization literature. The paper concludes that the model may be considered as an alternative to the 'Dixit-Stiglitz, iceberg, evolution, and the computer' framework which has served as the main tool-box in economic geography.