In this study I analyze the implications of contractual innovation in vertically-separated industries, using the example of the video rental industry. Prior to 1998, video stores obtained inventory from movie distributors using simple linear pricing contracts. In 1998, revenue-sharing contracts, which include inventory restrictions, were widely adopted. I investigate the effect of using revenue-sharing contracts on firms' products and consumer welfare, relative to linear pricing contracts. I analyze a new panel dataset of home video retailers that includes information on individual retailers' contract and inventory choices, weekly rentals and sales, and contract terms (prices and quantity restrictions) for 1,114 movie titles and 6,594 retailers in the U.S during each week of 1998 and 1999. A structural econometric model of rms' behavior is developed and estimated, and counterfactual experiments are performed. The results indicate that total upstream and downstream profits increase by three to six percent, and consumers benet substantially when revenue-sharing contracts are adopted. I also examine the effects of the observed quantity restrictions. I find that these restrictions serve to increase profits for upstream firms and decrease profits for downstream firms, relative to revenue-sharing contracts without inventory restrictions.