Measurement error in longitudinal earnings data: evidence from Germany

IF 1.6 Q2 INDUSTRIAL RELATIONS & LABOR

Journal for Labour Market Research Pub Date : 2024-06-04 DOI:10.1186/s12651-024-00366-x

Achim Schmillen, Matthias Umkehrer, Till von Wachter

{"title":"Measurement error in longitudinal earnings data: evidence from Germany","authors":"Achim Schmillen, Matthias Umkehrer, Till von Wachter","doi":"10.1186/s12651-024-00366-x","DOIUrl":null,"url":null,"abstract":"<p>We present evidence on the extent of measurement error in German longitudinal earnings data. Qualitatively, we confirm the main result of the international literature: longitudinal earnings data are relatively reliable in a cross section but much less so in first differences. Quantitatively, in the cross section our findings are very similar to those of Bound and Krueger (J Labor Econ 9:1–24, 1991) and Pischke (J Bus Econ Stat 13:305–314, 1995) for the United States while we find even stronger evidence that first-differencing exacerbates measurement error problems. We also show that measurement error in our survey data is not “classical” as it is negatively correlated with administrative earnings and positively autocorrelated over an extended period of time. Additionally, we estimate a model of measurement error stemming from underreporting of transitory earnings shocks in combination with a white-noise component and make a number of methodological contributions. Our results are robust to the use of two different linked survey-administrative data sets and various other sensitivity checks.</p>","PeriodicalId":45469,"journal":{"name":"Journal for Labour Market Research","volume":"2014 1","pages":""},"PeriodicalIF":1.6000,"publicationDate":"2024-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal for Labour Market Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1186/s12651-024-00366-x","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"INDUSTRIAL RELATIONS & LABOR","Score":null,"Total":0}

引用次数: 0

Abstract

We present evidence on the extent of measurement error in German longitudinal earnings data. Qualitatively, we confirm the main result of the international literature: longitudinal earnings data are relatively reliable in a cross section but much less so in first differences. Quantitatively, in the cross section our findings are very similar to those of Bound and Krueger (J Labor Econ 9:1–24, 1991) and Pischke (J Bus Econ Stat 13:305–314, 1995) for the United States while we find even stronger evidence that first-differencing exacerbates measurement error problems. We also show that measurement error in our survey data is not “classical” as it is negatively correlated with administrative earnings and positively autocorrelated over an extended period of time. Additionally, we estimate a model of measurement error stemming from underreporting of transitory earnings shocks in combination with a white-noise component and make a number of methodological contributions. Our results are robust to the use of two different linked survey-administrative data sets and various other sensitivity checks.

Abstract Image

查看原文本刊更多论文

纵向收入数据中的测量误差：来自德国的证据

我们提供了有关德国纵向收入数据测量误差程度的证据。从质量上看，我们证实了国际文献的主要结果：纵向收入数据在横截面上相对可靠，但在首次差额上就不那么可靠了。从数量上看，在横截面上，我们的研究结果与 Bound 和 Krueger（J Labor Econ 9:1-24，1991 年）以及 Pischke（J Bus Econ Stat 13:305-314，1995 年）对美国的研究结果非常相似，而我们发现更有力的证据表明，首次差分会加剧测量误差问题。我们还表明，我们调查数据中的测量误差并非 "经典 "误差，因为它与行政收入呈负相关，并且在很长一段时间内呈正自相关。此外，我们还估算了一个测量误差模型，该模型源于对过渡性收入冲击的少报和白噪声成分的结合，并在方法论上做出了一些贡献。使用两个不同的关联调查-行政数据集和其他各种敏感性检查，我们的结果是稳健的。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Journal for Labour Market Research INDUSTRIAL RELATIONS & LABOR-

CiteScore

2.60

自引率

5.90%

发文量

审稿时长

15 weeks

期刊介绍： The Journal for Labour Market Research is a journal in the interdisciplinary field of labour market research. As of 2016 the Journal publishes Open Access. The journal follows international research standards and strives for international visibility. With its empirical and multidisciplinary orientation, the journal publishes papers in English language concerning the labour market, employment, education / training and careers. Papers dealing with country-specific labour market aspects are suitable if they adopt an innovative approach and address a topic of interest to a wider international audience. The journal is distinct from most others in the field, as it provides a platform for contributions from a broad range of academic disciplines. The editors encourage replication studies, as well as studies based on international comparisons. Accordingly, authors are expected to make their empirical data available to readers who might wish to replicate a published work on request. Submitted papers, who have passed a prescreening process by the editors, are generally reviewed by two peer reviewers, who remain anonymous for the author. In addition to the regular issues, special issues covering selected topics are published at least once a year. As of April 2015 the Journal for Labour Market Research has a "No Revisions" option for submissions (see ‘Instructions for Authors’).