There is a consensus among academics and policy-makers that related lending, a widespread practice in most LDCs, should be discouraged because it provides a mechanism through which bankers can loot their own banks at the expense of minority shareholders and depositors. We argue that neither looting nor credit misallocation are necessary outcomes of related lending. On the contrary, related lending often exists as a response by bankers to high information and contract enforcement costs. Whether it encourages looting crucially depends on the other institutions that support the banking system, particularly those give depositors and outside shareholders incentives and mechanisms to monitor directors, and that give directors incentives to monitor one another. We operationalize this argument by examining an LDC banking system in which there was widespread related--Mexico from 1888 to 1913. We find little evidence, during this 25-year period, of tunneling or credit misallocation-even in the midst of a major, externally caused financial crisis that occasioned a government-organized rescue. The banking system was, in fact, remarkably stable and manufacturing enterprises that received related loans performed at least as well as their competitors.