Presumably enough Samizdatans read the financial reports of Pharma firms ad know when the patent life of said firms' money-spinning drug patents will expire. Most firms will/should try to offer an "advanced" version of that drug to spin out profitabillity of an expensively-researched and tested product. For all the drugs which fail in Phase II and IV clinical trials, and the costs this incurs - the effective patent life of 14 years, which Dr. Angell notes is an absolute minimum to keep research at reasonable levels.

It seems as though many people are confused about the details, including Dr. Posner. The 1994 law adopted by the US Congress in accordance with the Uruguay Round Agreements Acts switched from 17 years after the date of grant of the patent, to 20 years after the date of application of the patent. This was part of "harmonization" required by the Uruguay Round, as the majority of other countries used a system from the date of application. At the same time, there is a provision stating that if the process takes more than three years, then the patent life is extended by one day for each day over three years. The net effect is a patent life of at least 17 years after grant of patent, but no longer than 20 years. (And the latter only if the application ends up being quite quick.) It is not a three year extension.

Note that grant of patent can take place quite early in the approval process.

He is further wrong because there is no "further 5 year period for the time it takes for the FDA to approve." He is even more wrong because the recent 1999 amendment was about two things, one of which wasremoving a 5 year limit-- previously, term length extensions had a limit of 5 years. Thanks to the 1999 amendment, there is no limit on patent length extension. (Applicants can forfeit the extensions for a patent being stuck in an appeals process for not vigorously working to get it approved, though.) The second, and more important, is that the 1999 amendment introduced the principle of extending the patent length for the Patent Office taking more than three years in order to approve the patent. This in itself was not actually a major change; previously the law had been written so that the patent term was whichever was longer of the old "17 years from issue" and new "20 years from application" rules. However, this provision was determined to violate the WTO Uruguay Round provisions, which mandated the rule based on application instead of patent issue. Therefore, the Congress adopted a permissible solution based on extension if the Patent Office failed to approve the patent in a timely manner. The end result is the same, as you note.

Dr. Posner is completely wrong in suggesting that the patent time is extended based on FDA approval. It is extended based on Patent Office approval, not FDA approval-- but that was true under the old "17 years from patent grant" system.

See US Code Title 35 Section 154 here, and the notes on amendments here

I cannot speak to Dr. Angell's book or the data on effective patent life (and speed in the FDA approving drugs makes the most difference of all), but I do know what the actual changes in the patent law are, unlike Professor Posner.

Most firms will/should try to offer an "advanced" version of that drug to spin out profitabillity of an expensively-researched and tested product.

Yes, although it can seem ridiculous how much more people are willing to pay for slightly advanced "latest and greatest" drugs compared to what was state of the art two or three years ago. The majority of the time there is real improvement from the new derivative drugs, though I doubt that it's worth the cost. Tell it to someone who needs the drug, of course.

I am fully sympathetic to arguments that people spend too much trying to get tiny improvements from today's drug as opposed to the drugs of seven years ago, and I certainly agree that the pharmaceutical companies spend quite a bit of money to encourage such behavior. I can be persuaded that it would be better for behavior to change. However, I can't support any claim that such a (sensible) behavior change wouldn't affect research. Of course it would decrease research.

Comment deleted. Totally off-topic. Get lost

I think the current patent system is a good balance. It can really take a long, long time to get a new drug past the FDA and to market, and the expense is unbelievable. The FDA blithely heaps additional testing demands not required by statute onto company's shoulders, so if it weren't for a good, long period of patent protection, drug research would grind to a halt.

The real scandal is in copyright. It's been extended periodically this last century, and now stands at about 100 years if the exentions are applied for. Important works, culturally shared by everyone, remain the eternal properties of their owners.

Shouldn't Mickey Mouse, which EVERYONE in living memory grew up with, be public domain by now? Shouldn't the Beatles oldest songs hit public domain before the generation the hoisted them to fame dies of old age?

The purpose of copyright is just like patents--to encourage people to produce new work by giving them the monopoly to sell early on, but to allow the public to put these new ideas to use later on. The original copyright term of 25 years was a fair deal. Too much time actually discourages authorship by permitting companies and artists to recycle old stuff forever.

Shortening patent life might hasten innovation - the looming end of a patent may encourage the drug companies to bring to market their incremental advances faster.