Browsing by Subject "Pharmaceuticals"

The dissertation consists of three chapters relating to pricing strategies. Chapter 1 studies coupons for prescription drugs and their impacts on consumer welfare, firm profits, and insurance payments. Chapter 2 examines consumer brand loyalty and learning in pharmaceutical demand and discusses implications for marketing and health care policy. Chapter 3 develops a framework for estimating demand and supply in an online market with many competing sellers and frequent price changes and proposes optimal pricing strategies for a large seller.

The first chapter studies an innovative price strategy in pharmaceuticals. Branded drug manufacturers have recently started to issue copay coupons as part of their strategy to compete with generics when their branded drugs are coming off patent. To explore the welfare implications of copay coupons, I estimate a model of demand and supply using pharmaceutical data on sales, prices, advertising, and copayments for cholesterol-lowering drugs and perform a counterfactual analysis where a branded manufacturer introduces coupons. The model allows flexible substitution patterns and consumer heterogeneity in price sensitivities and preferences for branded drugs. The counterfactuals quantify the effects of copay coupons for different assumptions about the take-up of coupons and the ability of the branded manufacturer to direct them to the most price-sensitive types of consumers. The results show that the agency problem between insurers and patients gives a branded manufacturer a strong incentive to issue copay coupons. Introducing copay coupons benefits the coupon issuer and consumers but raises insurance payments. In equilibrium, insurer spending can increase by as much as 25% even when just 5% of consumers have a coupon, with social welfare falling significantly.

Medicines for chronic conditions like high cholesterol, heart disease, and diabetes are repeatedly used for a long period of time. Consumer dynamics thus plays an important role in the demand for those drugs. In the second chapter, I estimate a demand model with brand loyalty and learning using micro-level data from cholesterol lowering drug markets in the United States. The estimates suggest high switching costs and strong learning effects at the molecule level in the markets. Switching costs raise the predicted probability of choosing the same drugs in a row and learning largely increases patient stickiness to a molecule in the long run. I discuss pricing implications of the estimation results for drug manufacturers, insurance companies, and policy makers.

The last chapter, coauthored with Dr. Andrew Sweeting and Dr. James W. Roberts, looks at pricing in a different context. We estimate a model of entry, exit and pricing decisions in an online market for event tickets where there are many competing sellers and prices change frequently. We use the estimates from our model to analyze the optimality of the pricing policy used by the largest seller (broker) in the market. We show that the broker's pricing policies substantially affect the prices set by his competitors. When we compare the broker's pricing policy with the prices that our model predicts are optimal we find that the broker sets approximately correct prices close to the game, when his pricing problem resembles a static one, but that he might be able to gain from using different pricing rules and updating prices more frequently further from the game.

My dissertation explores the interactions between the various agents in the pharmaceutical industry and how they are affected by changes in health care policy. In my work, I examine innovation and competition among new brand drugs and the value of prescription drug insurance after patent expiration.

The second chapter of my dissertation empirically assesses the trade-off between patent breadth and patent length, a topic that has attracted significant theoretical but little empirical attention. I estimate a model of pharmaceutical demand and supply that incorporates insurance and advertising for the antidepressant market. Using these estimates, I consider the potential welfare effects of giving some of the most important product innovations broader but shorter patents, which increases the market power that these innovators have in the short-run but also allows for more rapid entry by generics. My results indicate that in this setting broader patents could increase total welfare by more than 9%, mostly through savings in insurer expenditures. These results are robust to endogenizing the entry of other branded drugs.

In the third chapter, which stems from research done jointly with Peter Arcidiacono, Paul Ellickson, and David Ridley, I use data from the pharmaceutical industry to estimate demand and supply for prescription drugs across both insured and uninsured consumers, allowing for consumer preferences organized into discrete types. I account for an important characteristic of health care markets: the price paid by insured consumers (copayment) is typically much smaller than the price received by the manufacturer. This analysis highlights how generic-drug availability differentially affects insured and uninsured consumers. In particular, generic entry disproportionately benefits insured consumers, at least in the first year to two years.

The fourth chapter in my dissertation extends the analysis in Chapter 2 to allow for a more generalized framework. In Chapter 2, the first pharmaceutical product innovation that enters a therapeutic class is assumed to be high-value while those innovations that follow are assumed to provide relatively little, if any, added therapeutic value beyond the first. Using the same data and demand model estimates, I consider the potential welfare effects of allowing these later to be considered high-value products and providing them with greater patent breadth and shorter patent length. My results indicate that in this setting, the modified patent policy could still increase total welfare by more than 8%, mostly through savings in insurer expenditures. These results are also robust to endogenizing the entry of other branded products.