Evaluation of the 'pull effect' of non-enforcement mechanisms

Debates on immigration control often contrast deterrence through restrictive policies with attraction through regularization. These expectations can obscure differences among policy instruments and recorded outcomes. This study examines the extent to which territorial exclusion, social exclusion, and regularization are associated with asylum applications and refused entries at external borders. Using a directed dyadic panel of 23 EU/EFTA destinations and 159 non-EU origins (2008–2019), we estimate Poisson pseudo-maximum likelihood models with dyadic and origin-year fixed effects. We compare composite policy indexes with their component instruments and distinguish current regularization route availability from magnitude-weighted cumulative reform history. The results describe conditional associations in recorded counts specific to pairs of origin and destination. We find that tighter territorial and social exclusion are associated with fewer refused entries. Broader regularization route availability is associated with fewer asylum applications, whereas a more liberalizing net reform history is associated with more. Specific policy instruments within the same composite index also exhibit divergent estimates. We demonstrate how aggregation and operationalization may shape conclusions about internal immigration control, especially regularization.