The paper makes a neat theoretical contribution. It derives the conditions that determine whether the skill distribution of the emigrants stochastically dominates (or is stochastically dominated by) the skill distribution of the stayers. Because the rewards to skills in Denmark are low (relative to practically all possible destinations), the model predicts that the emigrants will be positively selected, and that the skill distribution of the movers will stochastically dominate that of the stayers.
Our analysis of administrative data for the entire Danish population between 1995 and 2010 strongly confirms the implications of the model. Denmark is indeed seeing an outflow of its most skilled workers. And that is one of the consequences that a very generous welfare state must learn to live with.
The paper is forthcoming in the Economic Journal.