The patent system is full of tradeoffs. Maybe patents should only last for 3 years, maybe 10, etc. But for all its tradeoffs, patent protection does create an

The patent system is full of tradeoffs. Maybe patents should only last for 3 years, maybe 10, etc. But for all its tradeoffs, patent protection does create an incentive for massive investment in R&D. It's difficult to imagine a meaningful counterfactual scenario. Of course, bad patent office execution (leading to patent trolling, etc.) is parasitic and harms the system as a whole. But what about the FDA? The FDA is notoriously overly risk-averse when it comes to approving new treatments. It simply should not be hard to get a competing delivery mechanism approved. It's easy to test, the test of whether the mechanism works can be repeated at nearly zero cost thousands of times, offering substantially more data than the typical drug approval process yields (when the question is whether a drug is safe long term). If you are an FDA regulator, the only way you can really lose your job and pension is by approving a drug that should not have been approved. Nobody is keeping score of the suffering and loss of life caused by sluggish approvals.