[Research] Characterizing the International Migration Barriers with a Probabilistic Multilateral Migration Model

Author: 2016-10-31

研究成果:Xiaomeng Li, Hongzhong Xu, Jiawei Chen, Qinghua Chen*, Jiang Zhang & Zengru Di*, Characterizing the International Migration Barriers with a Probabilistic Multilateral Migration Model, Scientific Reports 6, 32522 (2016), DOI: 10.1038/srep32522.

 

国家间移民壁垒的量化及描述

 

简介:

 

人类迁移促进了现代文明格局的形成,并对各国经济、社会和文化的发展有重要影响。这一领域有很多值得研究的课题,其中最为基础的问题就是对“迁移原因”的探寻。个体选择理论是较为经典的、针对移民迁移模式的模型描述,其中迁移成本概念的提出,有其实际意义和研究价值。然而,由于该模型对个体迁移选择的描述限于目标国选择的确定性和双边性,使得该理论难以对目前国家间的移民壁垒进行描述和量化。

 

我们在个体选择模型的基础上,通过引入统计物理中的Boltzmann因子,构建了一个兼容多目标国家的概率迁移模型,并在此基础上,实现了对国际移民壁垒的量化和描绘。通过实证分析,我们发现了一些有意思的现象,也帮助我们对国家间人口迁移模式有了更为深刻的认识。

 

首先,两国之间的移民壁垒对于移入者和移出者呈负相关。简单而言,拥有较高移入成本的国家对移民的吸引度欠佳,其国民离开该国的意愿也更加强烈。其次,我们运用移民壁垒对各国进行聚类分析,试图就多边移民模式对各国进行分类,其结果呈现较为明显的地域和经济特性,也从这个角度验证了本文迁移模型的有效性。此外,我们还试图通过回归分析将移民壁垒进行分解,发现影响迁移成本的因素非常复杂,并通过GDP增长率、人均GNI、人类发展指数HDI等宏观指标实现了对迁移成本12.5%的解读。与早期的理论不同,我们发现近十年来,语言对于移民成本的影响作用已经很小,相对而言,移民网络的影响作用显著,即目标国特定族群的规模越大,就越能降低该类移民的迁移成本。

 

摘要:

 

Human migration is responsible for forming modern civilization and has had an important influence on the development of various countries. There are many issues worth researching, and “the reason to move” is the most basic one. The concept of migration cost in the classical self-selection theory, which was introduced by Roy and Borjas, is useful. However, migration cost cannot address global migration because of the limitations of deterministic and bilateral choice. Following the idea of migration cost, this paper developed a new probabilistic multilateral migration model by introducing the Boltzmann factor from statistical physics. After characterizing the underlying mechanism or driving force of human mobility, we reveal some interesting facts that have provided a deeper understanding of international migration, such as the negative correlation between migration costs for emigrants and immigrants and a global classification with clear regional and economic characteristics, based on clustering of migration cost vectors. In addition, we deconstruct the migration barriers using regression analysis and find that the influencing factors are complicated but can be partly (12.5%) described by several macro indexes, such as the GDP growth of the destination country, the GNI per capita and the HDI of both the source and destination countries.

 

原文链接:http://www.nature.com/articles/srep32522