This paper aims to determine an optimal allocation of the European Cohesion Fund
(ECF) and compares it with the observed ECF allocation. This optimal allocation
is the solution of a donor optimization problem who intends to maximize recipient
countries' GDP per capita to achieve economic convergence in the EU. It is shown
that the observed ECF allocation of the period 2014-2020 is different from our optimal
allocation. Besides, our solution identifies the recipient countries that can benefit from
higher ECF transfers than the observed levels as those having low relative GDP per
capita and good economic management. Poland and Romania are the 2 main winners
of the optimal allocation and this result is robust to changes in the specification of the
donor's utility function.