Thursday, 1 March 2012

Copyright: a proposal

As previously discussed, property rights for non-rivalrous goods are not the same as for tangible property, because they're not needed as a rationing mechanism to allocate finite resources.

However, we want new works to be created, and a good way to encourage people to do things is to pay them for it.  So we should keep copyright in some form.  But whereas copyright can enrich us by encouraging the creation of new works, it impoverishes us by restricting access to existing works.  We need to strike a balance.  The current Berne Convention stipulation of 50 years after the author's death is far too long - according to the EU it was intended to benefit two generations of the author's descendants.  The extension to 70 years after the author's death in the EU and the USA is supposed to allow for longer lifespans.  But it's fantastically implausible that a prospective author would be deterred from producing creative work by the consideration that her grandchildren might not receive royalties from it.

I suggest that we retain copyright in its current form, with the exception that we restrict it to five years from first publication.  That should be sufficient in most cases for the author to receive a fair return, but we should allow extensions year by year if the last year's sales of a work are the highest yet, so as to accommodate works that take some time to become popular.

This paper attempts a theoretical analysis of the optimum (welfare maximizing) copyright term, and comes up with fifteen years. Whereas this one, which I find more convincing, thinks two years is about right (reportedly the authors think their model unrealistic, and actually believe that copyright should not exist at all).  Before we adopt my proposal worldwide we should probably explore the optimum term more thoroughly: we might decide to adopt different terms for different classes of work.

I suggest that we add an additional right to protect one's reputation as the author of a copyrightable work, to apply during one's lifetime after the copyright period has expired.  This would mean that anyone using the work in public would be required to state whether it was used with the author's permission.  The author  would have the right to waive this requirement for some or all users.

Any objections?


  1. I am ashamed to say that the maths of this is beyond me, or it would take me a week to see if that is true. Instinctively the idea of writers in garrets or budding rockstars worrying about income streams 70 yeas after their deaths is absurd. However, is there a distinction between different forms? Eg a newspaper article might lose its value within a coupe of days, whereas a film might take several years to break even, while it goes from release, to DVD, to pay TV etc (let's us assume for now that movie accounts give true and fair view). Do these models take account of such different periods or do you reckon 5 years , with your reservation caters for that?

  2. The model parameters ought to be different for different sorts of work, so when the papers quote one term it's really just illustrative. The important point is that people discount future benefits, so extension of copyright terms beyond the first few years is going to evoke few additional creative works. If you accept that there is a trade-off to be made, it's very difficult to see how the optimal term can be many decades.

    1. Ah. I was worrying about complex maths and forgetting the easy stuff. Yes, even I can see (now) that small returns a long time in the future aren't much incentive. So I'm with you.

      Only problem is that it is you, John Kay, Kim Dot Com, one maths professor (linked on your blog) me (if I count) vs hollywood, record industry, academic publishers and maybe the entire software industry. Anyone else on our side?