Fairness in Matching Markets
Matching mechanisms play an important role in many areas of life, such as the allocation of seats in schools and universities, housing, or appointments at public offices. Many people consider allocation procedures that reward strategic behavior to be fair. This is the result of a study published in a WZB Discussion Paper by Tobias König (Linnaeus University), Lydia Mechtenberg (University of Hamburg), Dorothea Kübler, Director of the Research Unit Market Behavior at the WZB, and Renke Schmacker, Postdoctoral Researcher at the WZB. In an experiment, the researchers investigated which process participants choose when they are allowed to decide which procedure will be applied to other participants.
In many areas of life, matching mechanisms determine who gets which goods and services. Researchers have discussed various mechanisms that differ in whether they consider the priorities of applicants (e.g., admission criteria such as school grades, geographic distance, or lotteries) and whether they can be manipulated. Little is known about which methods people prefer and what features of a mechanism are considered fair. In a laboratory experiment, the researchers gave participants a choice between two mechanisms. On the one hand, they could choose the Boston mechanism, in which misreporting preferences can be individually advantageous and applicants can win out even if they have a lower priority than other applicants (e.g., poorer school grades). For example, the Boston mechanism is used in Berlin to allocate secondary school places. On the other hand, a mechanism could be chosen that cannot be manipulated and where allocation is strictly based on priorities. In the experiment, participants take on the role of impartial observers who decide which mechanism is used for other participants.
The study shows that many observers prefer the Boston mechanism. This is the case when applicants' priorities are based on entitlements earned through performance (i.e., meritocratically determined), and even more so when priorities are determined by lottery. At the same time, but especially when priorities are based on performance, there is also some support for the alternative mechanism that strictly considers priorities. Surveys of participants show that they do not attach great importance to the non-manipulability of a mechanism. Instead, a novel motive is observed that explains support for the Boston mechanism: even though non-strategic students suffer negative consequences, it is perceived as fair that the mechanism rewards smart strategic decisions. "Interestingly, many people find it fair for clever, strategic behavior to be rewarded. This could explain why in reality we often observe matching mechanisms that can be manipulated," says Dorothea Kübler.
The results are published in the Discussion Paper “Fairness in Matching Markets: Experimental Evidence”. The researchers suggest that in school choice, for example, if students' priorities are not based on merit but on lotteries, it might be advisable to consider a third mechanism, the so-called Top Trading Cycles (TTC) mechanism. Although this mechanism regularly defies students' priorities, it ensures efficient matching and cannot be manipulated.