数字平台经济正在崛起并迅速发展。很多公司如阿里巴巴、京东、滴滴等都创建了在线平台，这些平台促进并重塑了诸多商业活动。在线交易具有商品与信息相分离的特点，这一特点导致交易间的信息不对称问题尤为突出，本文聚焦于此，探讨电子商务平台上由信息不对称引发的的复杂管理问题。 本文首先关注零售类电商平台上商家与消费者之间的信息不对称问题。在线交易中高质量的商家往往希望消费者能够了解其真实的质量信息，因此本文探讨了一种目前在淘宝、京东等平台上新兴的创新型退货策略——退货运费险在传递商家质量信息方面的作用。退货运费险由保险公司承保，可以由在线零售商或消费者购买。在这种保险下，保险公司赔偿消费者在退货时所产生的运费。研究结果表明，运费险可以作为传递高质量信息的可靠信号，并且当消费者事前对整个市场缺乏信心且高低质量商品之间存在显著质量差异时，高质量零售商仅通过运费险就可以向消费者传递其高质量的信息；此外，通过京东平台上一万多家 零售商的数据，这一理论结果被进一步实证验证。与此同时，通过与免费退货这一退货策略进行比较，本文证明了第三方——保险公司的存在能够增强质量信号传递信息的能力。最后，本文发现由于信息不对称，引入退货运费险不一定能够使零售商受益，但却能惠及消费者，并且总能使保险公司获利。 本文进一步关注了在线平台与平台入驻商家之间的信息不对称问题。商家引发的不良事件（如销售劣质品）会给平台带来损失，商家的努力可以减少不良事件的发生，但关键在于平台如何为他们提供激励，一种可能的方式便是流量分配。 本文将该问题抽象为一个委托-代理问题，在一个连续时间无限时域并且风险中性的模型设定下，探索平台（即委托人）如何将固定数量的流量分配给一个或多个商家（即代理人）以激励其付出努力从而减少不良事件的发生。在单一代理人情形下，本文提供了具有解析形式的最优合同，该合同结构简单并且易于在实践中实施，同时，结果表明该情形下即使最优合同也无法实现资源的有效分配（即总是把所有资源都分配给代理人)；而对于多个代理人情形，有趣的是，本文发现资源的有效分配总能在一些激励相容的合同下实现，同时，本文设计了一个迭代算法来计算此类合同，并提供了一个具有解析形式且能够实现资源有效分配的最优合同。 本文的研究结果有助于缓解电子商务平台中的信息不对称性，提高在线市场的交易效率，并为平台运营管理带来新的管理启示。

The digital platform economy is emerging and growing rapidly. Companies such as Alibaba, JD, and Didi, are creating online networks that facilitate and reshape business activities. The characteristic of online transactions that goods separate from information results in strong information asymmetry. In this dissertation, I focus on complex managerial problems with information asymmetry on digital platforms. I firstly focus on information asymmetry issues on online shopping platforms between sellers and customers. High-quality online retailers always desire to convey their true quality to customers. Based on this, I examine the informational role of an innovative return policy, return insurance, emerging on various shopping platforms such as Taobao.com and JD.com. Return insurance is underwritten by an insurer and can be purchased by either a retailer or a consumer. Under such insurance, the insurer partially compensates consumers for their hassle cost associated with product return. I show that return insurance can be an effective signal of high quality. When consumers have little confidence about high quality and expect a significant gap between high and low qualities, a high-quality retailer differentiates itself from a low-quality retailer solely through its adoption of return insurance. I confirm, both analytically and empirically with a data set consisting of over 10,000 sellers on JD.com, that return insurance is more likely adopted by higher-quality sellers under information asymmetry. Furthermore, compared to free return (i.e., retailers directly compensate for consumers’ product-return hassles), return insurance is a stronger signal of quality, due to the role of the third party, the insurer. Despite its capability to signal quality, return insurance is costly for the retailer. Particularly, both high-quality and low-quality retailers are sometimes strictly worse off due to the option of purchasing insurance. Nevertheless, return insurance can improve consumer surplus and reduce product returns. Its profit advantage to the insurer is most pronounced under significant information asymmetry. In addition, I discuss information asymmetry issues between platforms and their product/service providers. There are adverse events (e.g., selling counterfeits) induced by providers that hurt the platform. Providers can exert effort to reduce adverse events. The question is how to incentivize them. One possible way is by allocating online traffic. I abstract this question as a principal-agent problem. Under a continuous-time infinite horizon and risk-neutral setting, I study how a platform (i.e., principal) should induce effort from one or multiple product/service providers (i.e., agents), who are more efficient in terms of welfare generation, by allocating a fixed amount of online traffic to reduce occurrences of adverse events. I formulate both a single-agent model and a multi-agent model as dynamic contract design problems. The analytical solution to the single-agent model is endowed with an intuitive structure and is easy to implement. The multi-agent model, on the other hand, is more complex to analyze. Surprisingly, I find out that efficient resource allocation (i.e., allocating all online traffic to agents) is always possible following some incentive-compatible contracts, and we devise an iterative algorithm to calculate such contracts. The results offer perspicuous economic explanations and insights. I also characterize in an easy-to-implement analytical form one of such incentive-compatible efficiency inducing contracts. This dissertation helps to alleviate the information asymmetry on digital platforms, enhance online market efficiency, and provide comprehensive managerial insights for platform operations.