A new paper published by the National Bureau of Economic Research analyzes the effects of “patent boxes,” special provisions found mostly in the EU that tax income from patented inventions at a lower rate.
A “patent box” is a term for the application of a lower corporate tax rate to the income derived from the ownership of patents. This tax subsidy instrument has been introduced in a number of countries since 2000. Using comprehensive data on patent filings at the European Patent Office, including information on ownership transfers pre- and post-grant, we investigate the impact of the introduction of a patent box on international patent transfers, on the choice of ownership location, and on invention in the relevant country. We find that the impact on transfers is small but present, especially when the tax instrument contains a development condition and for high value patents (those most likely to have generated income), but that invention itself is not affected. This calls into question whether the patent box is an effective instrument for encouraging innovation in a country, rather than simply facilitating the shifting of corporate income to low tax jurisdictions.
Practiced by 13 of the 51 countries studied, patent boxes are another way to subsidize innovation–or patent holders at any rate. What makes patent boxes uniquely interesting, the authors note, is the fact that they only subsidize ex post development (after the invention has been patented) rather than ex ante development (either through the promise of monopoly rents or directly subsidizing R&D).
The findings of the paper are that while patent boxes do increase patent transfers to countries with them, the effects are modest and do not appear to achieve the primary goal of these boxes, leading to a net loss in tax revenue for many jurisdictions.
[T]his change to the corporate tax systems did seem to increase the international transfer of patents into a jurisdiction, at least when there was no requirement for further development domestically, we found relatively little responsiveness overall. The choice of priority or applicant country was unaffected, and patented inventions did not increase in the countries offering a patent box. This last result is important, as it suggests that the primary stated goal of introducing a patent box has not been achieved.
The authors propose two potential motivations for patent boxes despite their limited benefits, especially compared to more effective alternatives for subsidizing innovation.
There are (at least) two arguments: the first (benign) one is that some patented inventions are produced with investment other than R&D but still have features that may create public goods in the form of information, justifying a subsidy relative to other investments. The second (less benign) one is that firms with commercially valuable patents are able to use some of their profits for rent‐seeking in the form of a reduced tax on some of their income. Put simply, a patent box subsidizes output rather than input, so it benefits mainly firms that have had success with their invention. This may in turn be an encouragement to all firms to undertake such invention, but it seems a fairly inefficient way to do so. Another disadvantage relative to R&D incentives is that such an instrument does almost nothing to alleviate the ex ante liquidity constraint faced by innovating firms (Hall and Lerner 2010).
One smaller finding of the study is that despite having a higher corporate tax rate during the time period examined (39%), the United States remains a “locus” of patent ownership, indicating that features of the US patent system make patenting in the US an attractive proposition for firms.