This paper studies the effects of price impact upon optimal investment, as well as the pricing of, and demand for, derivative contracts. Assuming market makers have exponential preferences, we show for general utility functions that a large investor's optimal investment problem with price impact can be re-expressed as a constrained optimization problem in fictitious market without price impact. While typically the (random) constraint set is neither closed nor convex, in several important cases of interest, the constraint is non-binding. In these instances, we explicitly identify optimal demands for derivative contracts, and state three notions of an arbitrage free price. Due to price impact, even if a price is not arbitrage free, the resulting arbitrage opportunity only exists for limited position sizes, and might be ignored because of hedging considerations. Lastly, in a segmented market where large investors interact with local market makers, we show equilibrium positions in derivative contracts are inversely proportional to the market makers' representative risk aversion. Thus, large positions endogenously arise either as market makers approach risk neutrality, or as the number of market makers becomes large.