Problem definition: Our research is motivated by the product expiration problem in consumer packaged goods retailing, which creates substantial landfill waste and drains firm profits. We analyze shipment policies (i.e., the rules to determine the quantity and age composition of inventory to ship from a warehouse to a retail location) and their impact on profits and waste. Academic/practical relevance: The same firm often bears the cost of expiration at the warehouse and the retail store, which is why the problem necessitates a supply chain perspective. The ship oldest first (SOF) policy (commonly referred to as first in, first out) is advocated by industry experts to manage product shelf lives. Although its optimality in a single location is well established in the literature, it has not been studied in the context of a two-stage supply chain. Methodology: We conduct empirical analysis on a real-life data set to motivate the relevance of our problem. Then, we formulate an infinite horizon dynamic programming problem with stochastic demand for which we obtain analytical and numerical results. Results: The SOF policy is found to always minimize waste at the warehouse and total waste (warehouse and retail level combined) and under certain practically unlikely conditions, to maximize profits. However, in most practical applications, it is suboptimal, and the optimal policy is shown to have a complex structure. We analyze deterministic and myopic versions of our problem in order to generate insights on the trade-off between the issuing cost and the expiration cost. Then, we develop heuristic policies based on the myopic analysis of the problem, which are shown to perform well in terms of profits, waste, and product freshness; in our numerical analysis, the best such heuristic yields a median optimality gap of 9.5% versus 21% for SOF, pantry life of 69% versus 56% for SOF, and retail waste of 4% versus 10% for SOF. Managerial implications: The SOF policy is shown to generate high waste at the retail store, where waste is more likely to be disposed of at landfills as opposed to being donated; therefore, it may have an adverse impact on the environment. Our results also show that it is not effective at managing shelf lives in the supply chain, contrary to what practitioners argue, as evidenced by poor pantry life leading to excessive waste at the household level. Our analysis also questions the value of flow-through stocking systems to facilitate SOF as we show that firms can gain much more from improving their issuing policies.