Almost every e-commerce site employs a so-called reputation mechanism that collects and publishes ratings from its users which then allow other market participants to make betterinformed choices. It is instructive to distinguish between two kinds of reputation mechanisms in accordance with the problems they address: those that are employed by online opinion forums, such as Amazon Reviews, are built to eliminate asymmetric information while those at online auction sites are primarily intended to induce cooperation and trust between the market participants [e. g., Dellarocas, 2006]. Consider the online auction site eBay as an example for this latter kind: its procedure is such that the winning bidder (henceforth the buyer) fifirst pays for the good and that the seller is required to send it only after receipt of payment. Without any trust-enabling mechanisms in place, the seller is best off keeping the good for himself1, even if he received the buyer’s payment. Since a rational, self-interested buyer can anticipate this, she will not pay for the good in the fifirst place and no trade takes place. While sanctioning reputation mechanisms, such as the one employed by eBay, address this problem, some of their game-theoretic assumptions are too strong for real-world marketplaces. Three of these assumptions stand out in particular: the assumption of truthful buyer feedback, the assumption of long-lived sellers and the assumption of sellers being incapable of whitewashing. In my thesis, I develop incentive-compatible trust mechanisms that do not require any of these assumptions. Furthermore, I focus on designs which avoid the strong common knowledge assumptions that prevented the application of previous proposals