<p>This paper constructs an agent-based model of a bank run by combining elements from two distinct agent-based modeling frameworks: fractional reserve banking and social networks. A single bank takes deposits from and makes loans to a large number of households. Households form social networks and choose to withdraw their deposits based on the actions of other households within their social networks via threshold modeling mechanisms. A liquidity event occurs if the bank has insufficient funds to satisfy all withdrawal requests. Simulation results suggest that the factors that most impact liquidity events are the type of threshold modeling mechanism employed, the degree of social reach, the incidence of social shifting, and the size of loans made.</p>