We study a two-echelon supply chain model consisting of a retailer and a manufacturer reacting to stationary i.i.d. stochastic consumer demand. Both the retailer and the manufacturer use the Order-Up-To policy to make replenishment orders. We derive the variance of the retailer’s order rate and inventory levels and the variance of the manufacturer’s order rate and inventory levels. We assume that costs in the supply chain are directly proportional to these variances (or standard deviations) and investigate the options available to the supply chain members for minimising costs. Our findings reveal that in the scenario where each supply chain member acts to minimise local costs, then the golden ratio turns up in the optimal gain in the inventory and WIP feedback loops of the retailer and manufacturer. However, if the players act to minimise global supply chains costs, then superior performance results. This is often at the expense of the retailer, but our results show that if the retailer takes responsibility for supply chain cost reduction and acts altruistically by damping his order variability, then the performance enhancement is robust to both the actual costs in the supply chain and a naïve or uncooperative manufacturer. Superior performance is achievable if firms coordinate their actions and if they find ways to re-allocate the supply chain gain.