This paper examines a two-stage supply chain where a retailer offers a return policy with partial refund to end customers, and a manufacturer offers a buyback contract for all unsold and returned items to the retailer to share the risk. Customer demand is stochastic and price dependent. Key performance measure is the utility of profit, a function of the mean and variance of profit. Supply chain members are risk averse. Optimal buyback price, wholesale price, and retailer's order quantity are determined for the supply chain under coordination. A computational study is conducted to investigate the impacts of the members' risk attitudes, refund amount, and retail price on the optimal decisions. Further analysis shows break-even points among utility values at different retail prices as the risk attitude parameters and refund amount change. These break-even points provide a guideline for the retailers to adjust the price to maximise the utility of profit.