A business school professor at NYU predicts that, without the option to sell back a game, not as many people would buy it new. Price Charting summarizes how the study's simulation of how the absence of a used market would impact the new game market:
Because gamers who buy new games no longer have the insurance to sell the game if they decide they don’t like it, they are less willing to take the risk of buying the game new. As a result, the elimination of a secondary market dampens new game sales.
Granted, the study relies heavily on a risk-aversive model, but it's a well-accepted one.