The U.S. Internal Revenue Service (IRS) makes publicly and freely available period migration data at the state and county levels. Among their uses, these data inform estimates of net-migration as part of the U.S. Census Bureau’s Population Estimates Program, which, in turn, are used for producing other annual statistics, survey design, business and community planning, and federal funding allocations. Building on and extending prior research, we devote this Research Brief to documenting from multiple new angles a highly concerning and apparently systemic problem with the IRS migration data since the IRS took over responsibilities for preparing these data from the U.S. Census Bureau in 2011. As we then discuss, despite the fact that the IRS provides documentation detailing changes that it made to how it prepares these data relative to how the U.S. Census Bureau prepared them, it is not clear why or how these changes would result in the problem detailed in our analysis. Given that this problem appears to be an internal one within the IRS, we conclude by suggesting that the post 2011–2012 IRS migration data not be used until this problem is resolved, and we encourage the IRS to do so quickly, transparently, and collaboratively.

美国国税局 (IRS) 公开和免费提供州和县级别的期间移民数据。在它们的用途中，这些数据作为美国人口普查局人口估计计划的一部分为净移民估计提供信息，而该计划又用于生成其他年度统计数据、调查设计、商业和社区规划以及联邦资金分配。在先前研究的基础上并加以扩展，我们致力于从多个新角度记录自 IRS 于 2011 年从美国人口普查局接管准备这些数据以来 IRS 迁移数据的高度关注和明显的系统性问题。然后讨论，尽管美国国税局提供了文件，详细说明了相对于美国如何准备这些数据所做的更改 人口普查局准备了它们，目前尚不清楚这些变化为什么或如何导致我们分析中详述的问题。鉴于这个问题似乎是美国国税局内部的一个问题，我们最后建议在解决这个问题之前不要使用 2011-2012 年后的美国国税局移民数据，我们鼓励美国国税局迅速、透明和协作地这样做.