David Kestenbaum of the Planet Money team had an interesting piece on whether patents are an impediment or spur to innovation. The immediate issue was the decision by Tesla Motors to put all its patents in the public domain with the hope of helping to create a mass market for electric cars. However the piece went further and asked the question of whether patents actually promote innovation.
The argument in the opposite direction is that they lock up technologies for the period of the patent's duration. They also create enormous legal uncertainties since the boundaries of a patent's applicability are rarely clear. This means that a deep-pocketed patent holder can often scare away potential innovators with the threat of a lawsuit.
The piece includes an interview with David Levine and Michele Boldrin, who have been warning of the economic harms of patents and copyrights for more than a decade. They also maintain the fascinating website AgainstMonopoly.org.
One area where I would disagree with their argument about experimenting with an alternative approach is the suggestion in the interview that the way to get from here to there is through a gradual shortening of patent duration. This may actually provide little benefit since all the legal structures around patents and the need for secrecy would still be in place. As a result there may be little, if any, perceptible benefit from reducing patent duration from 20 to 18 years, for example.
An alternative would be to carve out areas where research would be publicly funded with all findings and patents put in the public domain. For example, the government could set aside $5 billion a year to finance the research and development of new cancer drugs. We would then be able to compare progress in an area where the research is all openly available and the final products are all sold as generics compared to output in areas that rely on the current patent system. This would provide a quicker and simpler test of the relative merits of research supported by government granted patent monopolies as opposed to research supported by direct upfront funding.
(Publicly funded patents could be subject to a "copyleft" principle where anyone can freely use them as long as they themselves don't use the patent to develop a privately held patent. If they do go that route then they would have to negotiate a payment to the government just as they would a private patent holder.)
Note; Typos corrected.