[SleepyBob]

Originally Posted by Fiat_Lux

http://www.rufuspollock.org/economic..._copyright.pdf implies otherwise.

Amber

Actually, he doesn't. Or rather, he makes that an explicit assumption that creates that result. Models can only be as good as the underlying assumptions.

Lemma 3. Under assumptions 1.1 and 1.2 there exists a unique level of protection which
maximizes the production of creative work. We denote this by Sp. Furthermore, EITHER
there exists a finite solution to N0(S) = 0 and this is Sp OR no such solution exists and
Sp = infinity. With assumption 1.3 only the first option is possible.

Assumption 1.3 is that there is a point where increased copyright protection will decrease production. Remark 5 explicitly points out that without that arbitrary restriction, the optimal level may be infinite.

Remark 5. If the level of protection which maximizes the production of copyrightable
work, Sp, is infinite then no immediate statement can be made as to whether the optimal
level of protection, So, will be finite (and hence less than Sp) or infinite.

And once again, the period recommended by this paper and its formula is arbitrary. It merely creates a mathematical framework around its arbitrary goal and assumptions.

One example of a shortcoming of such a formula is that it only takes into account aggregate results. It is likely that for a number of popular authors, if they made $50,000 per book instead of $1,000,000 per book, they would be incented to write more, creating more "Welfare" to the world. A very, very short copyright would limit the amount of profit they could get from any one book, so they would be forced to continuously produce more books in order to make a living.

But, would it be right for us to prematurely remove their revenue stream from a profitable book in order to (in effect) coerce them to be more prolific, even if it did create more value for the rest of us? I would argue no, even if it did create more value for society as a whole.

All copyright terms are arbitrary in nature. You can argue that one term is "better" than another, but there isn't any that is uniformly better, and there will be winners and losers from any change. And the "best" length is entirely dependent on how you measure it. That's practically the definition of arbitrary.