Both House and Senate tax reform drafts include measures that are popularly being referred to as a US ‘patent box’. A patent box (also: innovation box; IP box) is in essence a special low tax rate applied to income generated by certain types of IP. As such, patent boxes may be viewed as high tax countries’ answer to tax havens – and even with a tax rate down to 20% the US would still be more a high tax country than a tax haven.
IP that has been moved to a tax haven has the potential to divert valuable tax revenue away from where the IP originated. Given the amount of valuable IP that is undeniably being created in the US – and the corresponding level of IP held in tax havens outside the US, this lost revenue must be an obvious concern. However, it wasn’t so much the concern for lost revenue as the concern for lost jobs that seems to be behind the patent box idea.
Why lost jobs?
Unlike tax havens, patent boxes are also supposed to stimulate innovation - although whether they actually do is a hotly debated issue among economists. At any rate, since the OECD’s 2015 BEPS initiative (Action 5), the idea is that there has to be a connection (‘nexus’) between the IP and the R&D that created it. And that is seen as a threat to US jobs. As the US Senate Finance Committee put it in their July 2015 report, the nexus principle ‘will have a significant detrimental impact on the creation and maintenance of intellectual property in the United States, as well as on the associated domestic manufacturing sector, jobs, and revenue base’.
How is the US responding to this concern?
The basic idea, which only features in the Senate’s proposals, would be a special tax rate for ‘foreign derived intangible income’ – think, in very simple terms: royalties received by a US corporation from outside the US. The rate is reached via a complex formula but comes down to an effective rate of 12.5% (increasing to around 15.5% after 2025).
In addition to this basic idea of a preferential tax regime for IP ‘exported’ out of the US, both House and Senate propose applying a special low rate of tax to ‘intangible income’ received by non-US subsidiaries, where this is not captured under existing anti-avoidance rules, such as subpart F (this income is euphemistically referred to in the Senate’s proposals using the acronym ‘GILTI’). These ‘GILTI’ rules would more or less neutralise the attraction for US based multinationals of using other countries’ patent boxes (or tax havens) – most of which would still be significantly below the proposed US rate.
So would these proposals save US jobs?
Maybe, but not for the reason given by the US Senate. Their understanding was that the BEPS nexus principle meant that the patent box tax benefit had to be conditional on the (R&D) activity being ‘substantially performed in the country where the discounted rate is being received’. So that would mean, in order to become BEPS compliant, foreign patent box regimes would start requiring the R&D to be located in their country rather than the US. That would certainly seem to be a potential threat to US jobs.
However, the final BEPS action 5 report on this only requires the R&D to be carried out by the taxpayer that gets the patent box benefits. There is not as such a geographical limitation. So it could be carried out through, say, a US permanent establishment. The only condition is that the permanent establishment should be taxable in the home country which grants the patent box benefits. This could certainly be a practical limitation to US-based R&D benefiting from certain foreign patent boxes. And if the IP generated from US based R&D was also taxable in the US that would likely cancel out the foreign patent box benefits. In practice countries may also limit the geographical scope of their patent boxes, the most recent example of which being Luxembourg, that will require the R&D to be carried out in the EU/EEA as from 2018.
Does the US patent box comply with the international rules?
So the concerns are real, but seem to call for more research. It is also the question whether the US Senate proposals themselves comply with the BEPS requirements, for example as regards what qualifies as IP: the current, implicit, US definition would cover a far wider spectrum of intangible property than permitted under the BEPS standard. It is also hard to see where the nexus principle features in the proposed draft legislation – albeit this would seem crucial to achieving its aim of preserving US jobs. And is the Senate’s additional proposal to grant tax relief for IP that is repatriated to the US intended to allow the subsequent IP income to benefit from the US patent box? If so this would be completely at odds with the BEPS prohibition on acquired IP getting patent box benefits. It’s also questionable how the GILTI proposal (or House equivalent) fits in with the BEPS rules: would it constitute a preferential IP regime? Should it be equated with related party outsourcing – which is in effect prohibited under the BEPS standard?
In short, a lot of questions that both legislators and taxpayers could usefully be asking themselves, or their advisors. Maybe the answers are in the yet to be issued implementing regulations. Or will the US simply disregard their international commitments on this? Until more is known, we should not expect a mad rush of scientists either into or out of Silicon Valley.