| Abstract | Due to the phenomenon that labor migration mostly happened within the same dialectal region, this paper tries to explore for the impact of dialectal distance on labors’ migration. Theoretically, we have proved that dialectal distance, with both identification effect and complementary effect, will first promote then prevent labors’ migration, under certain conditions. Empirically, based on Chinese Dialect Family Tree from ‘Dictionary of Chinese Dialect’ and ‘Language Atlas of China’, we have constructed pairwise dialectal distances of 278 prefecture, and a micro dataset of labors’ cross-dialect migration, by matching dialectal distances with ‘China Labor-force Dynamic Survey’(2012). With the micro dataset, we have verified our theoretical implications. Specifically, within the same dialectal region, when the cultural difference raise by one level, the probability of cross-prefecture migration will rise by 30%; however, across different dialectal regions, this probability will fall by 3%; and the optimal dialectal distance is across different dialectal areas but within the same dialectal region. Our finding is robust, and when we have controlled other factors such as economic factors, marriage and family, social network, Hukou system, geographic factors, and Mandarin, the inversed-“U”-shaped pattern still exists. |