Adverse Selection in the Marriage Market: HIV Testing and Marriage in Rural Malawi
中文摘要
如果伴侣质量随时间逐渐显现，婚姻市场中的信息不对称可能导致逆向选择并推迟结婚。性安全是一项重要但隐藏的伴侣属性，尤其是在HIV呈地方性流行的地区。一个同时包含可观察属性（吸引力）与隐藏属性（性安全）的正向同质匹配模型预测，消除关于性安全的信息不对称会使性安全的受访者更早结婚和怀孕；如果他们同时也具有吸引力，这一效应会更强。频繁的HIV检测可能使性安全的人得以传递信号并进行筛选。与这些预测一致，我们发现，一项高频、默认参与但可选择退出的HIV检测干预改变了人们对伴侣性安全的信念，并使结婚和怀孕提前：在28个月内，基线时未婚女性的结婚概率和怀孕概率分别提高了26%和27%。对于性安全且有吸引力的受访者，估计值更大。相反，单次检测干预没有这些效应，这与文献中其他HIV检测评估的结果一致。我们的研究结果表明，对HIV风险的内生反应可能解释为何HIV/AIDS的流行伴随着系统性的结婚和怀孕推迟。
Abstract
Asymmetric information in the marriage market may cause adverse selection and delay marriage if partner quality is revealed over time. Sexual safety is an important but hidden partner attribute, especially in areas where HIV is endemic. A model of positive assortative matching with both observable (attractiveness) and hidden (sexual safety) attributes predicts that removing the asymmetric information about sexual safety accelerates marriage and pregnancy for safe respondents, and more so if they are also attractive. Frequent HIV testing may enable safe people to signal and screen. Consistent with these predictions, we show that a high-frequency, "opt-out" HIV testing intervention changed beliefs about partner's safety and accelerated marriage and pregnancy, increasing the probabilities of marriage and pregnancy by 26 and 27 percent for baseline-unmarried women over 28 months. Estimates are larger for safe and attractive respondents. Conversely, a single-test intervention lacks these effects, consistent with other HIV testing evaluations in the literature. Our findings suggest that an endogenous response to HIV risk may explain why the HIV/AIDS epidemic has coincided with systematic marriage and pregnancy delays.