Indicator deep dive: ‘patent box regimes’ 

The Tax Justice Network updated its Corporate Tax Haven Index today with a new rolling batch. The index ranks countries by evaluating how much wiggle room for corporate tax abuse their laws provide and by monitoring how much financial activity conducted by multinational corporations enters and exits the countries. Laws are evaluated against 18 indicators. In this batch update, we focused on four of the indicators from the indicator group on ‘loopholes and exemptions’

In this blog, we’ll do a deep dive into one of these indicators: the ‘patent box regimes’ indicator.

Why penalise patent box regimes on the Index?

A patent box (often also called an intellectual property (IP) box or innovation box) is a corporate tax regime that allows companies to pay a significantly lower tax rate for a long term on income derived from patents and other qualifying IP.

A patent box is a profit-based tax incentive. At the Tax Justice Network, we generally believe that cost-based tax incentives should be preferred over profit-based incentives. Cost-based incentives (also known as expenditure-based incentives) reduce a company’s tax liability by allowing it to deduct investment costs from their taxable income faster or in greater amounts than the standard rules allow. Profit-based incentives, on the other hand, reduce the tax an investor pays on the profits their investment generates. These incentives reward commercial success and profitability, often creating windfall profits. Profit-based incentives also come with limited revenue cost predictability for government. In the case of cost-based incentives, the cost to governments is capped by the amount of investment.

Patent boxes are prime examples of profit-based tax incentives to stimulate Research & Development (R&D). In line with our general view described above, we however think cost-based R&D incentives like accelerated depreciation, investment tax credits and super-deductions are superior policies. For this reason, the indicator on patent box regime gives countries that have adopted a patent box (or a patent box like regime, as explained below) the maximum ‘haven score’ of 100.

Modifying the OECD’s nexus approach

Under the Indicator, countries that have a patent box but have added a ‘nexus constraint’ receive a slightly improved score of 90, compared to the score of 100 for a patent box without nexus constraint. The assessment of the nexus constraint is based on the OECD’s work under BEPS Action 5 on harmful tax practices. Under the OECD’s nexus approach, the royalty income derived from the IP can only benefit from a patent box regime in proportion to the share of qualifying R&D expenses (uplifted by 30%) on the total of the company’s expenses.

For example, a pharmaceutical company has developed a new drug and generates $1000 in profits from this drug. The company applies to benefit from the local patent box regime which offers a 10% rate on profits instead of the standard 25% rate of corporate tax. The patent box includes the OECD nexus constraint rule. To develop the drug, the company has spent $60 on in-house research (qualifying R&D expenses in light of the nexus constraint) and $100 on  acquiring external IP rights (not-qualifying expense). As a result, the ratio of qualifying expenses is $60/$100. Because of the uplift, $60 is increased by 30%, meaning that the final ratio is $78/$100. As a result, because of the nexus constraint, only 78% of the company’s $1000 profits can benefit from the preferential rate of 10% under the patent box.

The nexus constraint certainly limits the benefit of the patent box, but it does not prevent a situation where the bulk of qualifying profits already include significant tax benefits. In many countries, the application of the benefit is unrestricted in time, turning the benefit into a perpetual windfall gain. For this reason, we take into account the presence of the nexus constraint in countries’ patent box regimes, but we give it only a small reward.

To assess whether countries apply a nexus constraint to their patent box regimes, we rely mostly on the OECD’s Forum on Harmful Tax Practices (FHTP) peer reviews. But we also do a bit more digging. For some countries, the FHTP concludes that the nexus approach is in place, but the actual implementation of the nexus rules is riddled with loopholes. In Belgium, for example, the law implementing the innovation box includes the OECD’s nexus approach but adds that the proportion of qualifying income can be extended beyond the nexus limit if there is proof of exceptional circumstances which make it so that the additional value of the IP developing company is higher than the standard ratio. For our purposes, this kind of discretionary loophole (to define what are exceptional circumstances?) is sufficient to conclude that the nexus constraint has not been implemented correctly, even if the FHTP determines otherwise.

Patent box: (not really) a must for innovation?

Among the OECD countries that figure in the Corporate Tax Haven Index, about 65% have a patent box or a similar measure in place that exempts certain types of royalty income. Interestingly, on average, OECD countries with patent boxes rank about 9 places higher on the index than OECD countries without patent boxes or similar regimes.

These findings show two things. First, there is a clear pattern where OECD countries that embrace a patent box regime also tend to adopt other policies we consider to be prone to corporate tax abuse in the index. This clearly shows if we compare the average difference in haven score of OECD countries with patent boxes (average score of 63.4) and OECD countries without patent boxes (average score of 52). The haven score is the Index’ qualitative component, based on the performance of countries’ legal frameworks across all indicators (for more details, see the Index methodology). Adoption of a patent box generates a difference of 2.7 in haven score as compared to a country without a patent box. In that sense, the adoption of a patent box by an OECD country – which only accounts for a tiny fraction of the haven score – seems to be a good predictor for this country to have a bad overall score and thus of the country operating as a corporate tax haven.

Second, unlike what is often claimed in patent box countries, the group of OECD countries without patent boxes, show that the adoption of such a regime is not a prerequisite for a country to create a tax environment that incentivises innovation. Denmark, for example, has preferred to apply a super deduction for qualifying R&D expenses of up to 110%. Loss-making businesses can apply for a payment equal to the tax value of the negative income related to R&D. Instead of a patent box, Germany and Italy, too, grant innovators a tax credit of up to 25% and 20%, respectively, for qualifying R&D costs. The fact that Denmark, Germany and Italy are important hubs of innovation in the European region, indicates that the use of patent boxes is not necessary for fostering a climate for innovation. .

Patent box countries often defend their patent box regime by claiming that it not only incentivises local R&D but also prevents IP flight. Unlike in the case of cost-based R&D incentives, patent boxes are said to prevent the move of patents and other IP held by local companies to offshore companies of the same group once costs have been recuperated to benefit from other beneficial regimes elsewhere. It is true that in a globalised economy, it is relatively easy for multinationals to move intangible assets (IP) to low-tax jurisdictions. It is also true that under a patent box, companies have less of an incentive to move IP to low-tax jurisdictions as the patent box country itself has turned into one for income derived from the IP. However, since the adoption of the so-called DEMPE rules in transfer pricing (See Chapter VI of the OECD Transfer Pricing Guidelines), the justification of patent boxes as the only way to prevent IP flight has lost much of its validity. Under these new transfer pricing rules, legal ownership has largely become irrelevant for-profit allocation under IP. Instead, tax authorities now look at which group company does or has done the work on (D)evelopment, (E)nhancement, (M)aintanance, (P)rotection, and (E)xploitation of the patents and other IP in question. This means that countries can safely rely on cost-based incentive regimes without fearing future profits under the locally subsidised IP will be shifted to other group companies in other countries. Even if the ownership of the IP is moved to another group company and that group company will derive the income generated by the IP, the DEMPE rules will force the multinational to recognise that the IP was developed elsewhere in the group and adequate profit adjustment between the relevant companies will have to be made.

Patent box-like regimes: no-corporate-tax regimes and territorial tax regimes

Finally, our research reveals that a number of non-OECD jurisdictions have less visible but nonetheless aggressive tax policies in place to incentivise profit shifting through the hosting of IP in a local group company. Especially for externally acquired IP, these strategies remain relevant: they allow multinationals to buy existing patents and other protected immovable assets and locate them strategically in group companies in those countries where future royalty income is left untaxed. One such policy is for countries to exempt foreign income (including foreign royalties) as part of a purely territorial corporate tax system (i.e. a system that taxes only locally derived income). Such is the case in countries like Hong Kong, Liberia, Mauritius, Panama, Seychelles and Singapore A second obvious policy, and one which is embraced by many countries in the top 15 on the Corporate Tax Haven Index, is to simply refrain from levying any corporate tax in general. Logically, the absence of a corporate tax system has the same effect as a full royalty exemption without nexus. Such is the case in jurisdictions like Bermuda, the British Virgin Islands, Cayman Islands, Guernsey, Isle of Man, Jersey and Turks and Caicos.

Interestingly and worryingly, none of the countries adopting either of these strategies – exemption of foreign royalty income or absence of corporate tax – have had these policies marked as ‘harmful’ by the OECD’s FHTP. For countries without or with a zero corporate tax, referred to by the OECD as the ‘no or only nominal tax jurisdictions’, the FHTP has tried to translate its nexus requirements for patent box regimes (see above) so it can also be applied to the taxation of companies by no or only nominal tax jurisdictions. As such, under the FHTP’s rules, these jurisdictions should only allow companies to benefit from their zero corporate tax regime on taxable income (including income from IP) if the companies meet certain rebuttable substance requirements like having ‘adequate’ employees and expenditures. However, the FHTP’s tweak of the nexus constraint rule does not work very well. The nexus requirement for patent boxes is essentially a formula to determine how much income can benefit from a lower tax rate. The sanction for income that does not comply with the patent box nexus constraint is taxation at the ordinary tax rate. However, in no or only nominal tax jurisdictions, all income benefits anyway from a zero tax rate and thus there is no ordinary tax rate to fall back on as a sanction.

The FHTP’s prescribed alternative sanction is for the no-corporate-tax countries – all of them known corporate tax havens – to issue fines and eventually to strike the company that operates without substance off the register. In such case, the country should also spontaneously exchange all relevant information with the parent jurisdiction so the latter is aware of what has been going on with the local group company. It is however highly doubtful whether this ‘catch-22’ system of tax haven self-reporting can effectively work in practice: the parent jurisdiction will receive information from the tax haven, only if the tax haven itself agrees the local group company has acted as such. In any case, under the Patent Box Indicator, countries without or with a zero corporate tax receive the full score of 100, just like countries that have a patent box without nexus restraint.

Redefining harmful tax practices

Our research shows that there are quite a few lose ends on the current global standards for harmful tax practices in the field of R&D tax policies. The FHTP may deem patent boxes with or without a zero corporate tax as not harmful if they come with sufficient nexus constraint but this does not change the fact that many countries worry about base erosion through outbound royalty payments. Arguably, the rise of ‘defensive measures’ in the form of countries adopting anti-abuse measures in the form of limits to the deduction of royalties paid abroad is strong evidence that the FHTP mechanism to curb royalty flows to no-corporate-tax countries is insufficient: countries still worry about their tax bases being eroded because of outsized royalty payments to local group companies to group companies abroad where the group’s IP is held. As we have shown under the index indicator on ‘deduction limitation of royalty payments’, a rising group of countries has rules in place that disallow to some extent the tax deduction of royalty payments to related companies abroad.

Interestingly, among those countries with defensive measures restricting royalty deductions are also a significant number of non-OECD countries that are generally not considered tax havens. The fact that certain non-OECD countries worry about the state of global OECD shaped rules on harmful tax practices also corroborates with the most recent negotiation session of the UN Framework Convention, held in November 2025. One of the things discussed in Nairobi during these negotiations was the text of a commitment in the Framework Convention to address harmful tax practices. Probed by a few OECD countries who claimed the work done by the FHTP is perfectly adequate, Nigeria responded that the mere fact that the negotiators of the Framework Convention have received a mandate to do work on the matter of harmful tax practices is an acknowledgement that existing OECD standards on harmful tax practices are inadequate or unacceptable for some countries. In the view of Nigeria and other Global South countries, one of the goals of the Framework Convention should be to create a framework to deal with all harmful tax practices that works for all developed and developing countries in north, south, east and west.

Time will tell how alternative standards to fight harmful tax practices will take shape at the UN. It is clear that if the view of Nigeria and others would be turned into binding rules under the Framework Convention, this may result in a world without patent boxes and without countries that refuse to adopt corporate taxes. This, in turn, would prevent other countries from having to adopt defensive measures to deal with the negative spillovers of those tax practices that are currently deemed ‘not harmful’ but in FHTP name only.