Market integration drastically changes the interregional and international industrial specialization, which is accompanied by the expansion of interregional labor migration in terms of frequency, scale, and geographic distance. This process tends to divide the economy into urban and rural areas and to form an urban hierarchy. This pattern of globalization has a significant impact on the shape of the family. Are a parent and child more likely to live separately? If so, who will take care of the parents when they get old? Will the parents give their child a higher education if they expect the education to equip the child with sophisticated skills that will induce the child to migrate to some urban area faraway from the family's hometown? In this project I begin by elucidating how the background features of a region affect the patterns of a family's decision-making on the issues of migration, intergenerational living arrangements, education for the child by the parent, and care for the parent by the child, through empirical studies of the data from the U.S. and Japan. Next, I develop an economic geography model that explicitly incorporates the elucidated patterns of decision-making, and use the model to analyze how the decisions within a family and a region's position in interregional and international industrial specialization interact with one another and evolve. I also analyze whether the scope and depth of the market integration are constrained, and if not, how interregional and international industrial specialization is likely to evolve in the future. With my new economic geography model, I analyze how public policies in education and social security affect, and are affected by, interregional and international industrial specialization through the decisions made within a family. I also discuss which tier of government in the hierarchy of national and local governments should be responsible for these public policies.