Comparison of the Effects of Economic Differentials and Amenity Differentials on the Internal Migration in Post-war Japan

摘要本研究旨在比较20世纪下半叶日本区域经济差异与环境舒适差异、社会舒适差异对国内长途移民的解释力。正如我们之前有几篇关于国内移民的文章一样，我们做了很多基本的事实发现。我们假设人们的效用包括收入、环境便利和社会便利。我们主要使用两种方法；(1)利用修正的重力模型和净迁移模型进行回归分析；(2)对人口迁移原因进行全国和地级市调查。主要结果如下：“就业原因”一直是国内长途移民最突出的原因之一。“自然环境原因”在大多数调查中都没有被调查。虽然“家庭原因”在全国调查中占据了很高的比例，但这一原因并没有得到足够的重视。采用修正的重力模型和净迁移模型计算的结果如下：我们使用了10个主要地区之间的移民数据，使用了(1)1970年、1980年、1990年和2000年的全国人口普查和(2)来自1955年至2000年基本居民登记册的日本内部移民报告。首先，人均实际收入在20世纪下半叶总体上具有很强的解释力，但它一直在逐渐下降。其次，我们有一个非常有趣的发现，低平均气温促进了从寒冷地区迁徙到寒冷地区。然而，人们从寒冷地区向温暖地区迁移的明确证据还没有被普遍发现。第三，以人民生活指数衡量的地区生活水平差异对长途迁移的影响较弱。④年龄结构变化对各地区净迁移率变化的影响较弱。

The purpose of this study is the comparison of the explanatory power to internal long-distance migration in Japan of between regional economic differentials and amenity differentials such as those of environmental amenities and social amenities during the latter half of the 20th century. As we have a few preceding articles on internal migration, we made a lots of basic fact findings. We assume that people' s utility consists of income, environmental amenities and social amenities. We use two main methods; (1)regression analysis using modified gravity model and net migration model and (2)national and prefecutural survey of migration reasons.Main results are as follows."Employment reason" has been one of the most prominent reasons of internal long-distance migration. "Natural environment reason" has not been surveyed by most surveys. Though "family reason" have occupied high ratio on the national survey, this reason has not been given enough attention.The results using modified gravity model and net migration model are as follows. We use migration data between major 10 areas using (1)1970, 1980, 1990 and 2000 national census and (2)Report on Internal Migration in Japan Derived from the Basic Resident Registers from 1955 to 2000. First, per capita real income has generally strong explanatory power during the latter half of the 20^<th> century, but it has been declining gradually. Second, we got a very interesting finding that low average temperature promotes migration both from and to cold areas. The clear evidence, however, that people migrate from cold area to warm area was not found out generally. Third, regional differentials of standard of living measured by People' s Life Indicators affected long-distance migration weakly. Fourth, the effect of the change of age structure to the change of net migration rate of each area was found out to be weak.