Is Imputing Poverty Efficient? An example from refugee data in Chad Theresa Beltramo Hai-Anh H. Dang Ibrahima Sarr Paolo Verme1 December 2019 Abstract Collecting household survey data on refugees remains a challenge, at least in the foreseeable future, for various logistical and technical reasons. We address this challenge by applying cross- survey imputation methods to a combined survey and UNHCR census-type dataset to predict the welfare of refugees in Chad. Our proposed cross-survey imputation method offers poverty estimates that fall within a 95% margin of the true rate. This result is robust to different poverty lines, sets of regressors, and modelling assumptions of the error term. The method also outperforms widely used methods such as Proxy Means Tests (PMT) and the targeting method currently used by humanitarian organizations in Chad, although the latter performs surprisingly well given its simplicity. JEL classifications: C15, F22, I32, O15, O20. Keywords: Refugees; Forced displacement; Targeting; Poverty; Chad. 1 * Theresa Beltramo: United Nations High Commissioner for Refugees, Rue de Montbrilliant 92, Geneva 1201, Switzerland (e- mail:
[email protected]); Hai-Anh Dang: DECAT- Analytics and Tools Team-The World Bank, 1818 H St. N.W. Washington, D.C. 20433 (e-mail:
[email protected]); Ibrahima Sarr: United Nations High Commissioner for Refugees (e-mail:
[email protected]); Paolo Verme: Fragility, Conflict, Violence - The World Bank, 1818 H St. N.W. Washington, D.C. 20433 (e- mail:
[email protected]). The authors are grateful to Kristen Himelein and Aly Sanoh for helpful comments on a previous version. This work is part of the program ``Building the Evidence on Protracted Forced Displacement: A Multi-Stakeholder Partnership''.