| Abstract: | Men's and women's preferences are intercorrelated to the extent that men rank highly those women who rank them highly. Intercorrelation plays an important but overlooked role in determining outcomes of matching mechanisms. We employ simulation techniques to quantify the effects of intercorrelated preferences on men's and women's aggregate satisfaction with the outcome of the Gale–Shapley matching mechanism. Our results show that even a small amount of positive intercorrelation in a matching market means increased satisfaction for women and dramatically decreased potential for strategic manipulation. Negative intercorrelation has the opposite effects. Thus, matching markets characterized by positive intercorrelation are well suited for matching via Gale–Shapley, while markets characterized by negative intercorrelation may face opposition from the nonproposing side of the market. So that our results are immediately applicable, we also define and employ a general measure of intercorrelation that can be used for any matching market. |
