"Return migrants' self-selection: Evidence for Indian inventors"
Based on an original dataset linking patent data and biographical information for a large sample of US immigrant inventors with Indian names and surnames, specialized in ICT technologies, we investigate the rate and determinants of return migration. For each individual in the dataset, we both estimate the year of entry in the United States, the likely entry channel (work or education), and the permanence spell up to either the return to India or right truncation. By means of survival analysis, we then provide exploratory estimates of the probability of return migration as a function of the conditions at migration (age, education, patenting record, migration motives, and migration cohort) as well as of some activities undertaken while abroad (education and patenting). We find both evidence of negative self-selection with respect to educational achievements in the US and of positive self-selection with respect to patenting propensity. Based on the analysis of timedependence of the return hazard ratios, return work migrants appear to be negatively self-selected with respect to unobservable skills acquired abroad, while evidence for education migrants is less conclusive.
We study a variant of the capacitated vehicle routing problem (CVRP), which asks for the cost-optimal delivery of a single product to geographically dispersed customers through a fleet of capacity-constrained vehicles. Contrary to the classical CVRP, which assumes that the customer demands are deterministic, we model the demands as a random vector whose distribution is only known to belong to an ambiguity set. Moreover, we require the delivery schedule to be feasible with a probability of at least 1−ε, where ε characterizes the risk tolerance of the decision maker. We argue that the emerging distributionally robust CVRP can be solved efficiently with modern branch-and-cut algorithms if and only if the ambiguity set satisfies a subadditivity condition. We then show that this subadditivity condition holds for a large class of moment ambiguity sets. We derive efficient cut generation schemes for ambiguity sets that specify the support as well as (bounds on) the first and second moments of the customer demands. Our numerical results indicate that the distributionally robust CVRP has favorable scaling properties and can often be solved in runtimes comparable to those of the deterministic CVRP.