RUI: Geographic and Occupational Mobility During the Great Depression

在最近的金融危机、房地产市场崩盘和经济衰退之后，学术界对美国地理流动性的趋势，特别是经济波动时期的流动性行为，产生了越来越大的兴趣。越来越多的证据表明，美国的地理流动性呈下降趋势，经济低迷以多种方式加剧了这一趋势。最近的研究表明，自第二次世界大战以来，流动性在经济衰退期间通常会下降，而在过去几年里，负资产、抵押贷款锁定和财产税锁定等房地产市场因素对家庭流动性产生了强烈的负面影响。如果真的发生了“大衰退”，地理流动性出现周期性下降，这将揭示当前就业增长复苏的缓慢性质，并对政策制定具有重要影响。例如，自2009年底以来，职位空缺与失业工人的比率一直在上升。在正常时期，职位空缺数量的增加会伴随着失业率的下降，但在过去两年里，失业率基本上没有相应的下降。有一种假说认为，这可能是劳动力市场“错配”的证据：由于异常严重的经济衰退导致流动性异乎寻常地大幅下降，公司无法雇佣本来会搬家找工作的工人。确定失配在解释过去三次经济衰退，特别是大衰退中失业率缓慢复苏的重要性，是经济研究的一个重要持续主题。这项研究项目的目的是了解大萧条对20世纪30年代美国地理和职业流动性的影响。这一研究项目将通过探索大萧条对美国20世纪30年代十年间地理和职业流动的影响，有助于我们理解大萧条本身的经济学和历史，并有助于我们总体上理解经济波动对流动性的影响。为了实现这些目标，PI将与一组本科生研究合作者一起构建和分析丰富的新数据来源，使我们能够以全新的方式探索这些问题。目前对大萧条期间地理流动性的了解来自20世纪初的人口普查，这不是研究流动性和移徙的完美来源，因为它依赖对儿童出生地的观察来推断地点的变化；因此，仅限于有年幼子女的已婚男子。这一群体不能代表整个人口，也不能代表流动性最高的亚群：年轻人、单身人士和/或没有孩子的人。最近，新的数据来源和经验技术极大地提高了我们在历史时间框架内研究流动性的能力。这里提出的研究项目将利用这些新技术和来自1940年人口普查的新数据，这些数据已于2012年春季推出，以构建跨越大萧条十年和前二十年的关联面板数据集。

In the aftermath of the recent financial crisis, housing market bust, and recession, there has been increased scholarly interest in trends in geographic mobility in the U.S., and especially in the behavior of mobility during times of economic fluctuation. Mounting evidence points to a downward trend in geographic mobility in the U.S. exacerbated in multiple ways by the economic downturn. Recent work indicates that, since World War II, mobility has typically declined during recessions and that over the past several years housing market factors such as negative equity, mortgage lock-in, and property tax lock-in exert a strong negative effect on household mobility.If indeed the "Great Recession" has seen a cyclical fall in geographic mobility, it would shed light on the slow nature of the current recovery in employment growth and carry important implications for policy-making. For instance, since late 2009, there has been in increase in the ratio of job openings to unemployed workers. In normal times, an increase in the number of job openings would be accompanied by a fall in the unemployment rate, but for the past two years there has been essentially no associated fall in unemployment. One hypothesis suggests that this may be evidence of "mismatch" in the labor market: an inability of firms to hire workers who would otherwise move to find work, due to an unusually strong reduction in mobility rates brought on by the unusually strong recession. Determining the importance of mismatch in explaining the slow recovery of unemployment rates in the past three recessions, and especially the Great Recession, is a significant ongoing subject of economic research.The aim of this research project is to understand the effects of the Great Depression on geographic and occupational mobility in the U.S. during the 1930s. This research project will contribute to our understanding of the economics and history of the Depression itself and to our understanding of the effect of economic fluctuations on mobility generally by exploring the effects of the Great Depression on geographic and occupational mobility in the U.S. during the decade of the 1930s. To achieve these goals, the PI along with a team of undergraduate research collaborators will construct and analyze a rich new data source that will allow us to explore these issues in fundamentally new ways. What is currently known about geographic mobility during the Depression comes from the population censuses of the early twentieth century, an imperfect source for studying mobility and migration, as it depends on the observation of children's birthplaces to infer changes of location; and is therefore restricted to married men with young children. This group is not representative of the entire population, or of the sub-population with highest mobility rates: the young, single, and/or childless.Recently, new data sources and empirical techniques have dramatically increased our ability to study mobility in an historical time frame. The research project proposed here would exploit these new techniques and new data from the 1940 census that has become available in the spring of 2012 in order to construct linked panel datasets spanning the Depression decade and the two previous decades.