Patent Box Regimes in Europe

Patent box regimes (also referred to as intellectual property, or IP, regimes) provide lower effective tax rates on income derived from IP. Most commonly, eligible types of IP are patents and software copyrights. Depending on the patent box regime, income derived from IP can include royalties, licensing fees, gains on the sale of IP, sales of goods and services incorporating IP, and patent infringement damage awards.

The aim of patent boxes is generally to encourage and attract local research and development (R&D) and to incentivize businesses to locate IP in the country. However, patent boxes can introduce another level of complexity to a tax system, and some recent research questions whether patent boxes are actually effective in driving innovation.

As today’s map shows, patent box regimes are relatively widespread in Europe. Most have been implemented within the last two decades.

patent box regimes in Europe 2022 patent box Europe EU intellectual property tax regime IP

Currently, 13 of the 27 EU member states have a patent box regime in place. These are Belgium, Cyprus, France, Hungary, Ireland, Lithuania, Luxembourg, Malta, Netherlands, Poland, Portugal, Slovakia, and Spain (federal, Basque Country, and Navarra). Non-EU countries Andorra, San Marino, Switzerland, Turkey, and the United Kingdom have also implemented patent box regimes.

The reduced tax rates provided under patent box regimes range from 0 percent in San Marino to 12.5 percent in Turkey.

Italy repealed its patent box in 2021 and instead introduced a deduction for 230 percent of costs related to research and development. This represents a transition from a benefit based on income (the patent box) to a benefit focused on investment or expenditure (the super-deduction).

Table 1. Patent Box Regimes in Europe, as of August 2022
  Qualifying IP Assets Tax Rate Under Patent Box Regime Statutory Corporate Income Tax Rate
Patents Software Other (a)
Andorra   2% 10%
Belgium   3.75% 25.00%
Cyprus 2.5% 12.5%
France   10% 25.83%
Hungary   4.50% 9%
Ireland 6.25% 12.50%
Lithuania   5% 15%
Luxembourg   4.99% 24.94%
Malta   1.75% 35%
Netherlands 9% 25.8%
Poland   5% 19%
Portugal     10.5% 21%
San Marino (b)   0% or 8.5% 17%
Slovakia   10.5% 21%
Spain – federal (c)   10% 25%
Spain – Basque Country   7.2% 24%
Spain – Navarra   8.4% 28%
Switzerland (d)     Varies from canton to canton, up to a 90% exemption from corporate tax Varies from canton to canton; 11.9% to 21.6%
Turkey (e)     11.5% 23%
United Kingdom     10% 19%

Notes:

(a) “Other” refers to IP assets that are non-obvious, useful, and novel. These can only be applied to small and medium-size businesses.

(b) San Marino has two IP regimes. The “New companies regime provided by art. 73, law no. 166/2013” grants a tax rate of 8.5 percent. The “IP regime” grants a tax rate of 0 percent. Both apply to patents and software.

(c) The Spanish regions “Basque Country” and “Navarra” have separate corporate tax and therefore separate IP regimes.

(d) In 2020, Switzerland introduced a patent box regime at the cantonal level, which provides a maximum tax base reduction of 90 percent on income from patents and similar rights developed in Switzerland. Cantons can opt for a lower reduction.

(e) Turkey has a second IP regime which allows for a full tax deduction (0 percent effective tax rate) of qualified IP income resulting from R&D activities that were undertaken in Turkish Technology Development Zones.

Sources: OECD, “Dataset Intellectual Property Regimes”; Bloomberg Tax, “Country Guide”; PwC, “Tax Summaries”; EY, “Worldwide R&D Incentives Reference Guide 2022”; and OECD, “Tax Database: Table II.1. Statutory corporate income tax rate,” https://stats.oecd.org/Index.aspx?DataSetCode=TABLE_II1.

✔In 2015, OECD countries agreed on a so-called Modified Nexus Approach for IP regimes as part of Action 5 of the OECD’s Base Erosion and Profit Shifting (BEPS) Action Plan. This Modified Nexus Approach limits the scope of qualifying IP assets and requires a geographic link among R&D expenditures, IP assets, and IP income. To be in line with this approach, previously noncompliant countries have either abolished or amended their patent box regimes within the last few years.

Many European countries offer additional R&D incentives, such as direct government support, R&D tax credits, or accelerated depreciation on R&D assets. The effective tax rates on IP income can therefore be lower than the ones stated in the respective patent box regimes.

Tax Foundation Response to OECD Consultation on Amount A of Pillar One

Tax Foundation has engaged constructively in the OECD Pillar One and Pillar Two process since it was launched in 2019. The consultation on the Progress Report on Pillar One provides another opportunity for that. However, at this point it is challenging to be constructive when the policy seems designed to fail.

The Progress Report reveals that the ongoing technical work has continued to increase the complexity of the policy.

OECD Pillar One Amount A is meant to reallocate taxable profits of large multinationals, mitigate double taxation of profits, and avoid a harmful tax and trade war.

To accomplish those goals, though, the policy needs to be attractive for governments to adopt it, simple enough for agencies to administer it, and clear enough for companies to comply.

The cover note suggests that a multilateral convention to implement Amount A will have the force of law “only upon ratification by a critical mass of countries, which will include the residence jurisdictions of the ultimate parent entities of a substantial majority of the in-scope companies whose profits will be subject to the Amount A taxing right, as well as the key additional jurisdictions that will be allocated the obligation to eliminate double taxation otherwise arising as a result of the Amount A tax.”

The United States would need to ratify for the “substantial majority” phrase in the quote above to be satisfied. Ratification of such a tax agreement would need a two-thirds majority in the United States Senate—an extremely unlikely outcome.

The OECD Secretariat has clearly made progress in developing a policy that works in theory and may have a slight chance to work in practice. However, progress has not been made in developing a policy that achieves the goals of the project while making it possible that elected representatives in governments would be willing to adopt it.

These rules point toward bureaucracy and complexity rather than simplicity and transparency.

Early in the process one key criticism was that attempting to build a new structure on top of the existing set of international tax rules could create a frail superstructure. The frailty of the potential superstructure has not been resolved as more details of Amount A have come into focus.

It seems likely that rather than resolving challenges created by the current system, Amount A would introduce many new areas for disputes. The tax certainty processes and procedures become necessary and critical because this policy could cause disputes to increase. Despite attempting to use a formulary approach to allocate taxes and eliminate double taxation, the design lends itself to interpretations that will be disputed.

The complexity of the proposal is not lost on the authors of the document. Footnote 4 shows a small window into the challenges of policy. Referring to Title 5 Article 9: Allocation of the obligation to eliminate double taxation with respect to Amount A Profit, the footnote states, “Work is on-going to review and if necessary remove any logical inconsistency from the elimination framework.”

Logical inconsistency is thinking that this complicated policy can be readily adopted by governments across the globe.

Because the general policy is complex, it requires shortcuts for defining the scope of the tax base, sourcing revenues, and identifying the tax base that is being surrendered. But those simplifications themselves will lead to questionable (if not disputable) outcomes.

The process to identify an in-scope segment should include a measure of overall company profitability to avoid having a company that in general is in a loss position to calculate and pay Amount A.

When identifying the tax base, a formula is used to identify excess profits and then yet another formula to determine domestic profits that should be excluded. The formula seeks to identify excess returns. Excess returns are usually those above a normal return to an investment and that normal return being a profit above what the time value of money and inflation would provide.

As inflation has crept up across the globe, it has increased the importance of evaluating whether the metrics in Amount A should be referring to profitability margins after excluding a measure of inflation. A company that makes 15 percent profit where inflation is 2 percent should not be treated the same way as a company that makes a 15 percent profit when inflation is 10 percent.

Inflation adjustments should be widely applied throughout the proposal.

The safe harbor makes a critical error in not including withholding taxes in the calculation of what taxes have already been paid in a market jurisdiction. This could lead directly to double taxation and misses an opportunity to incentivize jurisdictions to adjust their approaches to source-based taxation.

The revenue sourcing approach provides multiple allocation keys, but it remains unclear whether the allocation keys are for general use or if the transaction-by-transaction approach is still preferred in most cases. As Tax Foundation has published before, the global allocation key for revenue sourcing seems more consistent with identifying large economies rather than indicators specifically related to cross-border trade.

The process of eliminating double taxation relies on formulas that appear to be without clear rationale other than providing a level of convenience for identifying which jurisdictions should be surrendering profits.

In summary, the rules reflect a high level of complexity alongside a handful of shortcuts that either reflect rough approximations or show weaknesses in the plan. This sounds not unlike the current system for allocating profits across borders.

New policies that will generate disputes are being developed and therefore new dispute resolution mechanisms are being designed.

It seems unlikely that this design will be an acceptable solution to governments as an alternative to existing complexity and disputes.

There may be an opportunity to salvage some of the project that is most likely to deliver more certainty and avoid disputes; perhaps Amount B, which could provide an agreed-upon benchmark for profits attributable to certain defined activities, would make that cut. However, a successful salvage effort would require keen attention to designing a package that would be acceptable to the governments that would have to administer and enforce the rules.