{"id":18698,"date":"2025-02-27T10:19:03","date_gmt":"2025-02-27T09:19:03","guid":{"rendered":"https:\/\/www.uni.lu\/fdef-en\/?post_type=news&p=18698"},"modified":"2025-02-27T10:20:33","modified_gmt":"2025-02-27T09:20:33","slug":"identifying-and-changing-the-impacts-of-migration-on-development","status":"publish","type":"news","link":"https:\/\/www.uni.lu\/fdef-en\/news\/identifying-and-changing-the-impacts-of-migration-on-development\/","title":{"rendered":"Identifying and Changing the Impacts of Migration on Development"},"content":{"rendered":"\n
\n

On 20 February 2025, we were delighted to welcome Prof. David McKenzie<\/strong> from The World Bank<\/a>.<\/strong><\/p>\n\n\n\n

As part of our lecture series on Cross-Border Labor Mobility, Prof. Ager delivered an insightful lecture on “Identifying and Changing the Impacts of Migration on Development”<\/strong>.<\/strong><\/p>\n\n\n\n

In his lecture, David began by noting that very many papers in the literature set out to estimate \u201cthe\u201d causal estimate of migration on some outcome. But in practice impacts are likely to be heterogeneous, depending on an individual or country\u2019s characteristics, the macroeconomic environment in which migration is taking place, and policy choices which shape these impacts. <\/p>\n\n\n\n

The lecture then highlighted how recent advances in two popular non-experimental causal inference methods: difference-in-differences, and shift-share instrumental variables, should be applied and interpreted once we take account of the heterogeneity in impacts. Three general lessons were that typically at best we can hope to identify some weighted average of impacts for units that have their migration status change in some assumed exogenous way:<\/p>\n\n\n\n

    \n
  1. First, we need to be clear in what is being measured, and note that it may tell us more about sudden or unexpected migration impacts than of more normal migration flows. <\/li>\n\n\n\n
  2. Second, careful use of these methods requires deep contextual understanding and making a rhetorical as well as statistical case for plausibility. <\/li>\n\n\n\n
  3. Third, inference is often more difficult than typically assumed, and our resulting estimates may be more uncertain as a consequence. <\/li>\n<\/ol>\n\n\n\n

    The lecture then turned from identifying impacts to discussing policy actions to change these impacts, explaining why spending on refugee flows vastly outnumbers that on increasing the benefits from economic labor mobility, and ideas for how to change labor mobility decisions when you are not able to change visa policies.<\/p>\n\n\n

    \n \n\"David <\/figure>\n\n\n


    This lecture series is organized by     Luxembourg Institute of Socio-Economic Research (LISER)<\/a> & the Doctoral School in Economics, Finance and Management from the University of Luxembourg, in the framework of the Fonds National de la Recherche (FNR)<\/a>, Luxembourg funded project ACROSS and financing scheme RESCOM.<\/p>\n\n\n\n

    The next lecture series will feature Prof. Sandra Rozo from The World Bank and will focus on \u201cInsights from Evidence on Colombia’s Regularization Programs for Venezuelan Forced Migrants<\/em><\/strong>\u201d.<\/p>\n\n\n\n