Do We Need New Theories of Irregular Migration?

Jasper Tjaden, University of Potsdam
Eduardo Acostamadiedo

Micro-level research on migration decision-making has traditionally focused on regular migration. However, with the recent surge in irregular migration in major destination regions (US, EU), there is renewed interest in understanding irregular migration as a distinct phenomenon. Systematic theory testing for irregular migration is still limited, leaving the mechanisms behind why some individuals select irregular migration poorly understood. This paper evaluates the predictive performance of classic migration theories for explaining irregular migration intentions, using the Todaro and Maruszko (1987) model as a benchmark. Following the rational-choice tradition, this model conceptualises migration as a trade-off between expected earnings at home, weighted by unemployment risks, and expected income abroad, weighted by unemployment risks and journey risks. The paper compares its predictive performance with other classic migrations theories, including income maximisation (Sjaastad 1962), network theory, and cultural norms (Massey et al. 1993). The article draws on survey data from over 10,065 respondents in Nigeria, Senegal, and The Gambia. The study applies multinomial logistic regression and machine-learning approaches to predict intentions to migrate irregularly. Predictive accuracy is assessed through ten-fold cross-validation using AUC-ROC metric, and variable importance is estimated with SHAP values. The study contributes to cumulative theory testing in migration research and provides empirical evidence on which theoretical mechanisms best account for irregular migration aspirations across different social groups.

See paper

 Presented in Session 87. International Migration Intentions