Association for Public Policy Analysis & Management

Patent protection often leads to high drug prices that make life-saving medicines less affordable to patients. To bring down prices, particularly in developing countries, government and generic drug firms increasingly infringe and invalidate patents. The situation is severe for treatments require multiple drugs owned by different firms with numerous patents, including drug bundling, or cocktails, for HIV. I study the impact of the first joint licensing organization for drug bundling (the Medicines Patent Pool) on global drug diffusion and innovation. To investigate this issue, I construct a novel dataset from licensing contracts, public procurement, clinical trials, and drug approvals. I find that, when the pool includes a drug for bundled licensing, generic firms worldwide can sublicense the drug cheaply and conveniently for sales in a set of developing countries, which leads to the supply of generic drugs increasing by 7% of the drugs in the country. The results are robust to alternative specifications, and the impacts are stronger in countries where a drug has extensive patents.

Firms respond to the pool with higher R&D input and firm participation in clinical trials, among both incumbents and new entrants. The reactions are stronger among non-branded firms outside the patent pool and concentrated on late-stage trials investigating new drug cocktails. In contrast, pool-participating firms re-allocate clinical resources to develop new drug compounds that can further complement existing ones. New drug product approvals increase primarily from faster generic approvals of existing drugs as well as new combinations of existing drug compounds. The increased generic approvals primarily benefit developing countries as they are not allowed to be marketed in developed countries when using licenses from the patent pool.

Finally, I structurally estimate a model of demand and supply to quantify the welfare gains and simulate counterfactuals. For the demand side, I estimate a nested discrete choice model, in which I utilize drug classification systems to construct nests for drugs with the same mechanism of action and allow cross-nest complementarity. For the supply side, I incorporate a simple model from the patent pool and bundling literature. My exercise aims to provide an illustration of the idea of drug cocktails in the context of the patent pool. I then simulate counterfactuals for potential benefits from future expansions of the pool.

Paper Linked Here.