College admissions with tuition transfers (Q2078089)

He introduces a college admission with tuition transfers problem. In this model, students' payments are not necessarily equal to their respective tuition fees. However, the total requested tuition fees must be equal to the total payment burden on students. In this model, he introduces two mechanism classes: ``deferred acceptance with efficiency-improving tuition transfers'' which is efficient, and ``deferred acceptance with stability-preserving and efficiency-improving tuition transfers'' which is stable and constrained efficient. Also, neither of them is strategy-proof. Next, for the Turkish college admissions, he introduces the mechanism ``deferred acceptance with limitedly efficiency-improving tuition transfers'' (DLT) that is stable, strategy-proof, and Pareto dominates the deferred-acceptance mechanism (DA), which is the best stable rule in the absence of tuition transfers. Lastly, he provides a welfare analysis to see how colleges are affected by tuition transfers. To this end, he considers a large class of college preferences and find that the most popular colleges may or may not prefer DLT to DA. In other words, their gain from switching to DLT from DA is ambiguous. On the other hand, all the other colleges unambiguously lose from that switch.