This estimate is also an overestimate according to the paper.
First, much creative endeavour builds upon the past and an extension of term may make it more difficult or costly do so. Were Shakespeare’s work still in copyright today it is likely that this would substantially restrict the widespread and beneficial adaptation and reuse that currently occurs. However we make no effort to incorporate this into our analysis despite its undoubted importance (it is simply too intractable from a theoretical and empirical perspective to be usefully addressed at present).
This means that the real number is significantly less than 15, maybe more like 12.
FaceDeer@fedia.io 15 hours ago
So glad to see another reference to this guy's work in the wild.
As an amusing side note, the original term of copyright in the first law that established it (the British Copyright Act of 1710) was a flat 14 years, with a mechanism that allowed you to apply for only one extension of an additional 14 years. So most things would be 14 years, and whatever select things were particularly valuable or important could have 28 years. Under Pollock's analysis this is just about the perfect possible system. So by sheer coincidence this is something that we got right the first time and ever since then we've been "correcting" it to be less and less optimal.