A country that experiences a shortage of workers with particular skills naturally considers two responses: import skills or produce them. Skill import may result in large-scale migration, which will not be to the liking of the natives. Skill production will require financial incentives, which will not be to the liking of the ministry of finance. In this paper we suggest a third response: impose a substantial migration admission fee, "import" fee-paying workers regardless of their skills, and use the revenue from the fee to subsidize the acquisition of the required skills by the natives. We calculate the minimal fee that will remedy the shortage of workers with particular skills with fewer migrants than under the skill "import" policy.
We ask which migration policy a developed country will choose when its objective is to attain the optimal skill composition of the country's workforce, and when the policy menu consists of an entry fee and a quota. We compare these two policies under the assumptions that individuals are heterogeneous in their skill level as well as in their skill type, and that individuals of one skill type, say "scientists," confer a positive externality on overall productivity whereas individuals of the other skill type, say "managers," do not confer such an externality. We find that a uniform entry fee encourages self-selection such that the migrants are only or mostly highly skilled managers. The (near) absence of migrant scientists has a negative effect on the productivity of the country's workforce. Under a quota: the migrants are (a) only averagely skilled managers if the productivity externality generated by the scientists is weak, or (b) only averagely skilled scientists if the productivity externality generated by the scientists is strong. In (a), a uniform entry fee is preferable to a quota. In (b), a quota is preferable to a uniform entry fee. If, however, the entry fee for scientists is sufficiently below the entry fee for managers, then migrants will be only or mostly highly skilled scientists, rendering a differentiated entry fee preferable to a quota even when the productivity externality is strong. Instituting a differentiated fee comes, though, at a cost: the fee revenue is not as high as it will be when migrants are only or mostly managers. We conclude that if maximizing the revenue from the entry fee is not the primary objective of the developed country, then a differentiated entry fee is the preferred policy.
