Manufacturer-sponsored copay assistance is an increasingly common and controversial practice in the US prescription drug industry. While manufacturers claim that their "copay coupons" help underinsured patients access valuable therapies, insurers are concerned that the discounts undermine cost-sharing as a means to address moral hazard and constrain prices. To date, research and policy on coupons have largely focused on coupons offered for branded drugs with generic equivalents. Much less is understood about coupons for branded drugs without generic equivalents, despite the fact that they comprise over half of all couponed drugs. In this paper, I develop an economic model of coupons and show that coupons may raise spending through both higher prices and higher quantities. Importantly, I also show that the effect of coupons on total welfare is ambiguous and depends in part on whether a generic equivalent is available. In a complementary empirical analysis, I combine a novel dataset of coupon offers with administrative claims data from a large national insurer to estimate the effects of coupon introductions between 2015 and 2018. I use difference-in-difference methods to address the potential endogeneity of coupon offers and introduce a measure of coupon exposure that exploits variation in cost-sharing across insurance plans. I find that for brands with a generic equivalent available, coupons increased prescription fills by 6%. However, coupons had limited, if any, impacts on utilization of brands that did not face generic competition. I also find evidence that coupons undermine the role of generic competition in constraining branded drug prices. Policy-makers should pursue restrictions on coupon use for branded drugs in the presence of generic equivalents and support further research on coupons in their absence.