 So I am an attorney and a faculty member at Harvard Medical School in the program on regulation therapeutics and law. At least 153 FDA approved drugs have received at least some public funding. Under the Buy Dole Act, these may be covered by patents, which can lead to high prices. There are several provisions within Buy Dole and outside it that can be used to limit patentee rights if it is in the public interest. I want to focus just on the March in Rights provision, which allows an agency to grant a license to a third party, such as a generic manufacturer, if the patentee has not achieved practical application of the invention where practical application is defined in the Buy Dole Act to require that the invention be available on reasonable terms. There have been six petitions for March in Rights, four of which were primarily focused on price. In these four, the petitioners generally asserted that prices were so high as to not be reasonable. The NIH denied all six petitions in the four in which price was the primary concern. The NIH took the position that practical application means that the drug is widely available and is not in short supply. Using the March in Prevision to reduce price could, says the NIH, affect investors' future willingness to invest and is therefore an issue that is appropriately left for Congress. The petitions pointed out that prices were substantially higher in the US than abroad. It was argued that these prices were unfair not only because they were much higher, but because US taxpayers had already paid for the initial research through NIH funding. That's the paid twice statement. The petitions pointed out that, sorry, assuming the government wants to act to lower prices, there are at least two complications in the case of FDA-approved drugs. First, drug companies have often developed a drug further and hold additional patents. These patents would, quote, not be subject to the government's March in Authority, as the NIH pointed out in its denial of the petitions for Latana, Prost, and Ratanavir. Second, FDA exclusivity means that patents are not the only barrier to the entry of generics. FDA exclusivity lasts five, seven, or 12 years, depending on the category of drug. And the Bidol Act does not directly address these exclusivity provisions. In addition, development times can be especially long with drugs. An important question is how much is left to do between the end of government funding and full market approval? If government funding only underwrites the cost of invention, say enough to identify the molecule and show in vitro activity with the tasks of formulation, animal studies, and especially human studies, left for a downstream drug company, then something may be needed to incentivize this development period. It doesn't need to be patents. It could be more government funding. Currently, the government's contribution may vary from case to case, but it appears to be small in relation to a drug company's cost structure. The amount of drug companies spent on research and development is relatively small, about 13.7% of revenues, according to one source. R&D can, in turn, be divided into clinical trials, meaning testing on humans and preclinical trials, which include the animal and laboratory testing. It's difficult to precisely identify how far government funding takes us, but generally, the central patents that cover drugs are filed before clinical testing begins, meaning that the government's or the university's contribution might be somewhere around here, potentially leaving substantial work left to do. On the other hand, in those cases, where government funding brings drugs closer to market approval, there's less need for maintaining private incentives. I want to finish with two policy suggestions that are not limited to the drug context. The first is that patents are often too long. The average number of words in a patent document has increased substantially since 1977. Patent attorneys are apparently trying to compete with law school professors for unnecessarily and counter-productively long documents. If you look at patents from the 1800s, including drug patents, some are only a page and a half long. Here we have ether, one of the most important drugs ever developed. It's less than two pages. Aspirin won a very few drugs to still be sold more than 100 years after it was introduced, one page. Vitamin E, less than two pages, smallpox vaccine, just over one page. Congress or the USPTO could increase fees for exceeding page limits. These are currently set at 400 for every 50 pages beyond the first 100 pages. Or for claims in excess of 20, currently at $80 per claim. Shortening patent length would reduce transaction costs, and it is transaction costs that underlie the ability of patent trolls to extract settlements on even weak patents. So second and relatedly, patent claims are often too vague or too indefinite in the lexicon of patent law. The problem is made much worse because claim terms by law are interpreted in view of the specification. If you have a 100 plus page specification, it's more likely that different portions of that document will support different and possibly contradictory interpretations of the claim language. Again, raising transaction costs, keeping Harvard Law School lawyers fully employed, and increasing uncertainty. The Supreme Court seemed to raise the bar for indefiniteness in 2012, but the general perception is that while the test has been modified, the bar has not moved much. So in addition to placing length limits on the patent document, the federal circuit could do more here to police the indefiniteness requirement.