1. These are the same authors behind the "Against Intellectual Monopoly" book, which, of course, they cite in this paper. In general, I would not trust anything these authors put out without fact-checking all their claims and actually reading the studies they cite. I have found that they are rather, uh, flexible with their facts: https://news.ycombinator.com/item?id=7889290

2. They claim they could find no evidence supporting a relation between patenting and innovation growth, only a relation between patenting and increased patenting. This is false - there are many studies showing positive correlation between patenting and measures of innovation other than patenting, such as R&D expenditure, and diversity of industries in which innovation happens. Just from a quick search:

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1513814

http://nw08.american.edu/~wgp/park_lippoldt08.pdf

http://www.dklevine.com/archive/sokoloff-kahn.pdf

http://catdir.loc.gov/catdir/samples/cam041/2003043048.pdf

http://www.nber.org/books_in_progress/invention/chapter7.pdf

http://www.nber.org/papers/h0042.pdf?new_window=1

Heck, patenting is also positively correlated with increased prosperity in general:

http://www.brookings.edu/~/media/research/files/reports/2013...

3. The bit about the correlation of patenting and productivity growth is also incorrect, and is actually something of a strawman. The theory is innovation results in increased productivity, and patenting is a proxy for innovation, but the relation between innovation and productivity is ambiguous and difficult to quantify:

http://www2.itif.org/2013-competitiveness-innovation-product...

http://elsa.berkeley.edu/pub/users/bhhall/papers/BHH11_Innov...)

But even then, a quick Google suggests that there is evidence of patenting and increased productivity: http://cep.lse.ac.uk/people/bloom/BloomVanReenen.pdf

The authors certainly are aware of Google, so I'm guessing they simply decided to cherrypick studies that support their point of view.

4. Of course, there are costs imposed by the patent system as well, and there are a number of studies showing these as well, many of which this paper cites. The big question these days is, how can we tune the system optimally so that the benefits outweigh the costs? To get a more nuanced view of the situation, see this metastudy:

"RECENT RESEARCH ON THE ECONOMICS OF PATENTS", Bronwyn H. Hall Dietmar Harhoff