Weather variation and temporary labor migration: a panel data analysis for select semi-arid villages in India. (3rd May 2020)

• Record Type: Journal Article
• Authors: Pradhan, Kalandi Charan; Narayanan, K.
• Is Part Of: Migration and development. Volume 9:Number 2(2020)
• Page Start: 291
• Page End: 322
• Publication Date: 2020-05-03
• Subjects: Temporary migration -- weather variation -- demographic factors -- socio-economic factors -- India; Emigration and immigration -- Periodicals; Emigration and immigration -- Economic aspects -- Periodicals; 304.82
• DOI: 10.1080/21632324.2019.1605745
• Languages: English
• ISSNs: 2163-2324

Abstract

ABSTRACT: The purpose of this study is to understand the relationship between weather variation and temporary labor migration. In doing so, we investigate how labor migration is used as an adaptation strategy to weather variation for the select Indian semi-arid villages. We use crop yield deviation as a proxy variable for the weather variation at the household level. In order to investigate the objective of the study, we employ panel data of 210 households using Village Dynamic South Asia (VDSA) data set of six villages from the state of Telangana and Maharashtra for the period 2005–2014. We have used village, state and aggregate level logistic regression models to demonstrate how factors at each of these levels can influence temporary labor migration trajectories. The study finds that the score of crop yield deviation for the non-migrant households is higher as compared to migrant households, which shows that migrant households have a higher adaptive capacity to weather variation as compared to their counterpart households. Further, the empirical evidence from the logistic model shows that along with demographic and socio-economic factors, weather variation influences the temporary migration at aggregate level (full sample estimation). In addition to this, we also find that weather variation is statistically significant in determining temporary migration for entire Maharashtra at state level estimation and only one village (Kalman) at village level estimation.