The Tax Attractiveness of EU Locations for Corporate Investments: A Stocktaking of Past Developments and Recent Reforms

3.1.1 Tax Rate Cut and “Mini-IRES” Regime

A common measure to attract corporate investments is a reduction in the statutory corporate tax rate. To give an example, France has continuously decreased its corporate income tax rate from 33.3 percent in 2019 to 25 percent in 2022. In addition, the French legislator decided to gradually abolish the Cotisation sur la Valeur Ajoutée des Entreprises (CVAE) ^[16] to increase the competitiveness of French companies.^[17] Figure 3 shows the effect of the French reform on the EATR. The decrease in the statutory corporate tax rate and the CVAE translate into a decline in the EATR of 7.8 percentage points (from 32 percent to 24.2 percent).

Figure 3:

Impact of a tax rate cut and the “Mini-IRES” regime on the EATR (in percent). Notes: The figure compares EATRs of corporations before and after incorporating the reform options discussed in Section 3.1.1 in percent. While the reform is still pending in IT, FR already implemented the tax rate cut in the period from 2019 to 2022. IT: best (worst) case refers to a 100 (zero) percent qualifying investment in the Italian reform proposal. Source: Authors’ contribution.

As opposed to a general corporate income tax rate cut, Italy decided in favour of a system with a regular tax rate (IRES ^[18]) and a reduced tax rate on certain qualifying investments and new hires (“Mini-IRES”).^[19] The aim of the reform, which was approved in 2023, is to stimulate economic growth and ensure greater international tax competitiveness.^[20] While the specific design of the regulation is not yet known, some elements might be adopted from a similar reform proposal in the 2019 budget law.^[21] For example, the definition of qualifying investment is expected to align with the “Transition 4.0” initiative^[22] of the Italian government, incentivising companies that invest in the digital transformation of production processes.^[23] The reduced tax rate is supposed to precede the execution of the investments, i.e. companies are subject to the reduced tax rate in the first place and in case they do not perform a qualifying investment, they have to pay back the difference between the regular rate and the reduced rate in subsequent periods. Although the specific tax rates are not yet known, the reduced tax rate is expected to align with the global minimum tax rate of 15 percent.^[24] When modelling the reform, we assume that the regular tax rate remains at 24 percent. Figure 3 depicts the EATRs for two extreme scenarios, i.e. 100 percent qualifying investments (“best case”) and zero percent qualifying investments (“worst case”). In case of qualifying investments only, the reduced tax rate of 15 percent would result in an EATR of 16.1 percent. In contrast, the EATR would remain at 23.6 percent if the firm did not perform any qualifying investment. However, in reality, the effective tax burden might lie between these extreme values.

In sum, Figure 3 illustrates that the corporate income tax rate is the main driver of the EATR. This can be attributed to the fact that the EATR measures the tax burden on a highly profitable investment, i.e. the profit tax has a large impact compared to other non-profit taxes and tax base regulations.

3.1.2 Accelerated Depreciation

As mentioned in Section 2, countries also rely on accelerated depreciation schemes to boost economic activity. To highlight the impact of this measure on the EATR and thus on countries’ tax attractiveness as investment location, we focus on two countries implementing different accelerated depreciation schedules.

As part of the Growth Opportunities Act,^[25] the German legislator plans to temporarily introduce a declining-balance depreciation with a rate of 25 percent instead of a linear depreciation scheme for tangible assets.^[26] Figure 4 shows that the impact of this reform proposal on the EATR is only marginal.^[27] The small decrease of 0.2 percentage points results from the fact that in the model applied for the calculation of the effective tax burden, the more favourable depreciation scheme would apply to machinery, which represents only 20 percent of capital employed. However, even in case of a 100 percent investment in machinery, the EATR-reducing effect would increase to 1.2 percentage points (from 28.3 percent to 27.1 percent) only. This is due to the fact that the current German legislation allows companies to depreciate machinery linearly over its useful life. Assuming a useful life of seven years, the difference between the status quo and the reform proposal is relatively small. A larger reduction in the EATR may only be achieved with more generous depreciation rules as recently implemented in the United Kingdom.

Figure 4:

Impact of accelerated depreciation schemes on the EATR (in percent). Notes: The figure compares EATRs of corporations before and after incorporating the reform options discussed in Section 3.1.2 in percent. While the reform is still pending in DE, the UK already implemented the reform in 2023. Source: Authors’ contribution.

In 2023, the United Kingdom’s corporate income tax rate was increased from 19 percent to 25 percent, but at the same time full expensing, i.e. a 100 percent capital allowance on qualifying plant and machinery investment, was implemented on a temporary basis. Prior to the reform, machinery had to be depreciated on a declining-balance basis at a rate of 18 percent. Figure 4 displays the impact of both reform elements separately. The six percentage point increase in the statutory tax rate raises the EATR from 20.1 percent to 25.6 percent. However, when also taking into account the possibility of an immediate write-off of machinery, the EATR decreases to 24.4 percent. Taken together, both measures have opposing effects on the effective tax burden.

3.1.3 Notional Interest Deduction

Another policy option that is currently discussed is the introduction or even harmonisation of notional interest deduction regimes. As outlined in Section 2, some EU Member States already apply notional interest deductions. However, the notional interest rate and the specific design of the regime differs across countries. In 2022, the European Commission presented the so-called Debt-Equity Bias Reduction Allowance (DEBRA) proposal that aims at reducing tax-induced distortions between equity and debt financing.^[28] One of the proposed measures is the implementation of an allowance on equity using a (currency-specific) harmonised notional interest rate. If the DEBRA proposal is adopted, countries that already have a notional interest deduction regime in place will have to apply DEBRA’s provisions instead of their domestic regulations.^[29]

Figure 5 illustrates the heterogeneous impact the DEBRA proposal has on Member States’ effective tax burdens based on two example countries. While Spain currently does not apply a notional interest deduction and represents a high-tax country, Poland has a very generous regime in place with a notional interest rate of 7.75 percent. For the simulation of DEBRA, we use an interest rate of 4.09 percent.^[30] The implementation of the aforementioned measure would decrease the EATR in Spain by four percentage points, while the EATR in Poland would increase by 2.2 percentage points. Thus, depending on the generosity of the current national regime, the DEBRA proposal can have heterogeneous effects on the effective tax burden. Compared to the accelerated depreciation schemes shown above, the introduction of an allowance on equity has a larger impact on the EATR. However, both effects are strongly dependent on the exact parameters, i.e. the depreciation rates, notional interest rates and the respective asset and financing weight.^[31]

Figure 5:

Impact of notional interest deduction regimes on the EATR (in percent). Notes: The figure compares EATRs of corporations before and after incorporating the reform options discussed in Section 3.1.3 in percent. In both countries, the reform is still pending, as the European Commission has not yet implemented DEBRA. Source: Authors’ contribution.

3.1.4 R&D Tax Incentives

In addition to general measures reducing the effective tax burden, countries grant tax incentives for specific activities. In the context of the so-called “smart tax competition”, R&D tax incentives are a widely used measure to attract mobile activities. A distinction between input- and output-based incentives has to be made: Input-based incentives refer to more generous tax deductions for R&D expenses (in the form of tax credits, super-deductions and accelerated depreciation for assets used in the R&D process), while output-based incentives grant reduced tax rates for income resulting from the innovation process (so-called patent box regimes).

To examine the impact of these R&D incentives on the effective tax burden, some model assumptions have to be modified.^[32] More specifically, we consider a self-developed patent instead of an acquired patent in all of the following R&D incentive simulations. This allows us to model incentives related to different R&D activities within a company, such as expenses for personnel, machinery and buildings. Before analysing the impact of R&D tax incentives on the EATR, we highlight the effect of the modified model assumptions. While an acquired patent has to be capitalised and depreciated over several years, expenses for a self-developed patent can usually be immediately expensed. To illustrate the effect of these different depreciation schedules, Figure 6 contrasts the EATRs resulting from an investment in an acquired patent versus a self-developed patent in selected countries.^[33] While the EATRs of these two types of assets are relatively similar in Germany, Italy and the United States, the effective tax burden from the investment in a self-developed patent is much lower compared to the investment in an acquired patent in Spain, France and the United Kingdom. These heterogeneous effects can be explained by the different lengths of the underlying depreciation schedules for acquired patents in the aforementioned countries.

Figure 6:

EATRs in the baseline scenario and in the R&D scenario without incentives (in percent). Notes: The figure compares EATRs of corporations in the baseline scenario and the R&D scenario without incentives (in 2023) in percent. In the baseline (R&D) scenario, we assume an acquired (self-developed) intangible asset. While the acquired intangible is capitalised and depreciated over several years, the self-developed intangible can usually be immediately expensed. Source: Authors’ contribution.

In the following, we analyse the impact of R&D incentives on the EATR. Figure 7 displays the results of several sensitivity analyses with respect to changes in the generosity of R&D incentives. On the one hand, it illustrates the impact of incentives compared to the ordinary tax burden and on the other hand, it compares the R&D scenario in the status quo with different reform options.^[34]

Figure 7:

Impact of R&D tax incentives on the EATR (in percent). Notes: The figure displays EATRs of corporations with and without R&D incentives before and after incorporating the reform options discussed in Section 3.1.4 in percent. While the reforms are still pending in DE and IT, the UK (US) already implemented the reform in 2023 (2022). Source: Authors’ contribution.

Although Spain and France have a relatively high ordinary effective tax burden (see Section 2), they provide generous R&D tax credits with a broad tax base and a rate higher than the corporate income tax rate as well as a patent box regime. Moreover, they allow for an accelerated depreciation of tangible assets used for R&D. Figure 7 shows that due to the aforementioned R&D incentives currently in place, the EATR in both countries is reduced by around eight percentage points. Thus, countries can increase their attractiveness significantly by implementing generous R&D tax incentives.

While R&D incentives also have been in place in the remaining countries depicted in Figure 7, they either were recently amended or are planned to be modified. In 2020, Germany introduced a tax credit for eligible R&D expenses amounting to 25 percent. Currently, eligible expenses include R&D personnel expenses only. However, the German legislator plans to extend the assessment basis to expenses for movable fixed assets used for R&D activities. This expansion would decrease the EATR by 1.3 percentage points and thus make R&D investments more attractive. However, the overall impact of R&D incentives on the effective tax burden is rather limited, as apart from the tax credit, there are no other incentives in place.

In contrast to Germany, the United Kingdom provides various R&D incentives such as an accelerated depreciation for qualifying assets, a patent box regime granting a reduced tax rate of ten percent for income from intellectual property and an R&D tax credit. As part of the aforementioned reform in the United Kingdom that increased the corporate income tax rate and introduced a more generous depreciation schedule, the R&D tax credit was increased from 13 percent to 20 percent. Figure 7 illustrates that although the tax credit becomes more generous, the overall effective tax burden increases. Thus, the effect of the tax rate increase from 19 percent to 25 percent outweighs the effect of the more generous tax credit. However, without the tax credit amendment the EATR would be even higher.

As mentioned before, most countries allow for an immediate expensing of R&D personnel expenditure, even if they do not grant any further incentives. While in the United States in the past, researchers’ wages could be immediately expensed in the year they were incurred, since 2022, these expenses are required to be capitalised and depreciated over six years. Although the same amount of expenses can be deducted from the tax base in both cases, the immediate recognition of expenses is more favourable for companies in terms of interest and liquidity effects. Particularly in times of rising interest rates, the effect would be similar to an interest-free loan from the government, which could support investment activity. This can also be seen in Figure 7, as the EATR increases from 22.5 percent to 23.7 percent in the capitalisation scenario.

Finally, in the context of the implementation of the “Mini-IRES” regime and against the background of the global minimum tax, the Italian legislator plans to revise and simplify all tax incentives. It is expected that the current tax incentives are gradually replaced by the application of the reduced tax rate for certain investments. To examine the impact of a potential shift from a tax-incentive system to a tax system based on a reduced tax rate for qualifying investments, we assume that all R&D incentives are abolished. Instead, R&D activities are regarded as qualifying investment and therefore subject to the reduced tax rate. Under the assumption that – as before – R&D investment accounts for 20 percent of the overall investment and the remaining 80 percent are not regarded as qualifying investment, we compute a weighted average tax rate of 22.2 percent.^[35] In addition, we do not include any R&D incentives in the simulation. Under the status quo, the EATR including the R&D incentives in place is relatively low (14.4 percent) compared to the case without incentives (23.6 percent), suggesting that the incentives are generous. In contrast, under the application of the “Mini-IRES” regime without any tax incentives, the EATR increases significantly to 22.1 percent. However, this estimate may rather represent an upper bound of the EATR, as the percentage of qualifying investments is likely to be larger than 20 percent in many cases.

3.1.5 Overall Comparison of Tax Reform Options

After having examined different reform options separately, we compare their concrete potential for improving EU Member States’ tax attractiveness for corporate investments. Figure 8 provides a comprehensive overview of the effective tax burden in the status quo as well as under the aforementioned reforms.^[36] Overall, the implementation of a tax rate cut (as illustrated for Italy and France) can increase the attractiveness of a country significantly. The implementation of the “Mini-IRES” regime in Italy limits the application of a reduced tax rate to returns on qualifying investments only. While an accelerated depreciation scheme does not have a large effect on a country’s position in the EATR ranking, the introduction of a notional interest deduction – as suggested by the DEBRA proposal – reduces the EATR significantly.^[37]

Figure 8:

EATRs under status quo and with reform options (in percent). Notes: The figure displays EATRs for corporations in 2023 in percent. The EATRs of countries marked with an asterisk (dark coloured bars) incorporate the reform options discussed in Section 3.1. In case of FR and the UK, the dark coloured bar represents the pre-reform status and the light coloured bar the status quo. Source: Spengel et al. (2024)/Authors’ contribution.

In sum, tax rate changes and notional interest deduction regimes have the largest impact on the tax location attractiveness. In contrast, more generous depreciation schemes only have a minor effect as they usually are only applicable to specific assets. However, the impact on the effective tax burden strongly depends on the specific design of the respective measure.

3.2.1 Tax Rate Cut and “Mini-IRES” Regime

As has been shown in the development of EATRs over time (see Section 2), the statutory tax rate is a strong determinant of the EATR. This illustrates the important role of tax rate reductions in the international tax competition for mobile economic rents, i.e. to attract the inflow of foreign direct investment (or prevent its outflow). Moreover, in terms of tax policy, a reduction of the statutory corporate income tax rate has a major signalling effect due to the high salience of the corporate income tax rate. Against the background of numerous anti-avoidance regulations, the location choice of multinational firms will even more depend on the statutory corporate tax rate as ex-post shifting of profits to lower-tax jurisdictions becomes more costly.

This is in line with the finding of Dobbins and Jacob (2016) that prior to the comprehensive anti-avoidance initiatives, domestic firms reacted more strongly to tax rate cuts than foreign owned firms (which were able to shield considerable parts of their profits from high tax rates by relocating them to low-tax jurisdictions). Further studies indicate that a reduction in corporate income tax rates is associated with higher levels of innovation (Atanassov and Liu 2020; Falck, Kerkhof, and Pfaffl 2021) as well as an increase in investments, total wages, private consumption and GDP (Dorn et al. 2021).

The advantage of tax rate cuts comes at a cost. Most prominently, tax rate cuts imply immediate revenue losses that are considerably larger compared to alternative tax incentives that materialise for new investments only (Dorn et al. 2021). In addition, even for large corporate tax rate cuts the effect on economic growth and investment might be less substantial than that of more specific incentives (Hanappi, Millot, and Turban 2023).

With the implementation of the global minimum tax in the EU starting from January 2024, the EU limits the scope for corporate tax rate cuts. Statutory tax rates below 15 percent become less attractive as they would likely result in effective tax burdens below this threshold and thus trigger the application of a top-up tax (Ferreira Liotti et al. 2022). However, it is important to note that this minimum tax is applicable solely to large multinationals that have revenues surpassing EUR 750 million. Therefore, significant reductions in tax rates remain a viable option for small and medium-sized businesses. In this context, it is also evident that general tax rate reductions are more appealing to countries with high tax rates as in most cases they still have room to decrease statutory tax rates towards the global minimum tax rate of 15 percent.

An alternative scheme of a tax rate cut distinguishes between favourable tax rates for certain qualifying investments and a general rate. Generally, the proposed introduction of the “Mini-IRES” regime (as quantified for Italy in Section 3.1) is better suited to incentivise investments, as only those firms that actually invest benefit. This idea is operationalised in two ways: In the first approach, companies pay the regular corporate income tax to the tax administration, and receive subsequent tax refunds for certain promoted investments. In the second approach, as recently proposed by Italy, a more favourable tax rate is initially applied to all companies, in a way similar to a general tax rate reduction.^[38] However, if companies fail to meet the pre-specified investment requirements, they have to pay back taxes.

As only those companies that actually invest benefit from the lower rate, the “Mini-IRES” regime encourages investment in a more targeted way and also prevents windfall gains. It is a compromise in the sense that it limits the revenue loss for the government. Yet, the benefits primarily comprise those of a tax cut. The second approach, in which the lower tax rate is applied beforehand, effectively constitutes an interest-free loan to companies. In this scenario, companies should have more liquidity at their disposal for investment since lower tax payments occur before potential investment activities. However, the administration of such a system is more complex compared to a single corporate tax rate as companies have to substantiate their eligibility to the reduced tax rate.^[39] An increased complexity of the tax system can be negatively related to FDI (Hoppe et al. 2020; Lawless 2012). Again, as for the general tax rate cut, the global minimum tax as well as revenue constraints limit the applicability of this policy.

3.2.2 Accelerated Depreciation

Beyond tax rate cuts, tax base measures, such as accelerated depreciation are regularly introduced as investment stimulus (recent examples include Germany and the United Kingdom, see Section 3.1). In these accelerated depreciation regimes, the depreciation allowances are concentrated on the early years of an asset’s economic life while the deductions from the tax base are reduced accordingly in later years. Thus, it does not lead to a long-term loss of tax revenue, but rather in a postponement of tax payments into the future (Dorn et al. 2021). This makes it attractive from a government perspective. The tax benefit for the investing firm results from the so called timing effect of taxation: early tax savings yield a value if the interest rate is positive. Again, the postponement of tax payments essentially represents an interest-free loan for companies (Domar 1953).

Several empirical studies corroborate the effectiveness of accelerated depreciation empirically. For example, Eichfelder, Jacob, and Schneider (2023) and Maffini, Xing, and Devereux (2019) find that accelerated depreciation schemes lead to higher levels of investment. These findings are in line with the model calculation of Dorn et al. (2021), where accelerated depreciation has positive effects on the level of investment as well as on wages, consumption and GDP. However, the investment effect size depends on the existence of liquidity constraints. Zwick and Mahon (2017) find a higher investment response if accelerated depreciation results in an immediate cash flow due to tax savings in the same period as opposed to a postponed cash flow (e.g. when a firm is loss making and tax advantages only materialise in the future).^[40] From the government’s perspective, in the best case scenario, the lower tax revenues will be offset by additional revenues due to successful investments and the resulting increase in economic activity (Dorn et al. 2021; Goode 1955).

Potential side effects of accelerated depreciation include opportunities for tax planning, increased complexity and compliance costs. Opportunities for tax planning, for example, arise due to the presence of multiple, overlapping incentives linked to one asset (Hanappi, Millot, and Turban 2023). In addition, despite the stimulating effect of accelerated depreciations, the quality of investment does not necessarily increase. In fact, Eichfelder, Jacob, and Schneider (2023) find a decrease in the quality of investments resulting from these types of tax incentives because they diminish the marginal cost of capital. Consequently, while in total more investments are carried out, a part of them is profitable only due to the additional incentive.

As mentioned above, accelerated depreciation schemes allow for higher deductions in the early years of an asset’s economic life, which could result in an effective tax rate below 15 percent and thus trigger a top-up tax. However, when calculating the effective tax burden for global minimum tax purposes, not only actual taxes paid but also deferred taxes are taken into account. Deferred taxes arise from temporary differences between the actual tax expense and the tax expense according to financial accounting and thus also include accelerated depreciations for tax purposes. In the context of the global minimum tax, accelerated depreciation schemes or immediate expensing can therefore still be used as a measure for stimulating investment without increasing the risk of a top-up tax (Ferreira Liotti et al. 2022).

3.2.3 Notional Interest Deduction

In our analysis, the measure that results in the sharpest decline in EATRs after the tax rate reductions presented is the introduction of a generous notional interest deduction. This measure aims at creating an investment-promoting environment as it shields the marginal return of the investment from taxation. By allowing an additional allowance to be deducted from the tax base that reflects the cost of equity financing, a notional interest deduction addresses the debt-equity bias that arises from the distinct tax treatment of debt and equity (Devereux and Freeman 1991). Thus, it helps to achieve two primary objectives: enhancing the investment environment for domestic corporations and attracting foreign direct investments (Konings, Lecoq, and Merlevede 2022).

In the context of our study, one of the major benefits of notional interest deduction regimes lies in their neutrality regarding marginal investment decisions, as these regimes ultimately result in the taxation of economic rents only (De Mooij and Devereux 2011; van Campenhout and van Caneghem 2013). Furthermore, the notional interest deduction remains neutral for various asset classes, such as tangibles or intangibles, as the deduction is applied in principle and is only linked to equity (Konings, Lecoq, and Merlevede 2022). With respect to the widespread debt-equity bias, a notional interest deduction decreases the cost of equity and contributes to a more equal treatment of debt and equity.^[41] The IMF (2016) emphasises that removing the debt bias (from external borrowing) helps to ensure macroeconomic stability. Empirical studies corroborate this positive impact of a notional interest deduction on reducing debt levels of firms (Hebous and Ruf 2017; Romaniuk and Malik 2021).

The empirical evidence on investment effects of notional interest deductions is mixed. While Hebous and Ruf (2017) do not find significant effects on production investment after the implementation of the notional interest deduction in Belgium, Konings, Lecoq, and Merlevede (2022) show positive effects on employment and investment. Domestic Belgian firms experienced an increase in their after-tax return on investment due to the notional interest deduction (Konings, Lecoq, and Merlevede 2022).

However, the deduction of the cost of equity financing increases the complexity of the calculation of the tax base as well as the risk of cross-border tax planning (Hebous and Ruf 2017). Furthermore, the tax benefit might be quite sensitive to the overall interest environment. From a government perspective, a notional interest deduction regime could be quite costly, as the narrower tax base ultimately reduces tax revenues collected. Finke, Heckemeyer, and Spengel (2014) derive a revenue loss of up to 18 percent from a microsimulation model for Germany. According to their simulations this could be financed by an increase of the corporate profit tax rate by six percentage points to compensate for the narrowing of the tax base. The costs of a notional interest deduction regime could be capped, if one limits the benefit only to new equity which is brought into the company.

As illustrated in Section 3.1, a notional interest deduction reduces the effective tax burden significantly. Although the effect size depends on the notional interest rate and the amount of a firm’s equity, such a regime makes the application of a top-up tax more likely. As the notional interest can be deducted from the tax base continuously and is not available under financial accounting rules, it results in a permanent difference between both sets of rules. Therefore, in contrast to an accelerated depreciation scheme, a notional interest deduction is not covered by deferred taxes and decreases the effective tax rate by all means. In sum, the effectiveness of such a regime is likely to be reduced by the application of the global minimum tax if it results in an effective tax burden below 15 percent (Gschossmann et al. 2024).

3.2.4 R&D Tax Incentives

Our quantitative analysis shows that R&D tax incentives are an important instrument to enhance the tax attractiveness of countries in the competition for (R&D) investment. These tax instruments are used by legislators to encourage investment in specific industries or innovative markets (Hymel 2006). In particular, for investment activities where the benefits are highly uncertain and public spill overs are larger than private ones, the optimal level of investment is not undertaken by the private sector (e.g. R&D, green transition, digitalisation). However, higher R&D investment increases economic growth as measured by GDP (Akcali and Sismanoglu 2015). In addition, persistent R&D performers seem to survive crises better than their competitors (Lome, Heggeseth, and Moen 2016), suggesting that R&D investment acts as a form of insurance against future economic downturns.

R&D tax incentives thereby influence both corporate decision margins, the extensive one (as reflected by the EATR) and the intensive one. First, generous R&D incentives increase the post-tax net present value, indicated by a decrease in the EATR. This increases the attractiveness of the location for (new) R&D investments. Second, these incentives reduce the cost of capital, which leads to additional investments as more projects prove profitable from an investor’s perspective. Empirical evidence confirms that if input-based R&D tax incentives are sufficiently generous, they do increase R&D investment (Appelt et al. 2016; Guceri and Liu 2019; Hall and Van Reenen 2000; OECD 2020). This increase in investment and employment leads to more output (Lerche 2022) and, consequently, higher tax revenues for governments. Yet, increases in R&D investments may also be due to the relocation of R&D activities and not an overall increase in R&D activity, i.e. R&D tax regimes have competitive effects (Knoll et al. 2021). In contrast, there is only limited evidence on the encouraging effect of output-based R&D tax incentives on innovation activity within a country (Alstadsæter et al. 2018; Gaessler, Hall, and Harhoff 2021).

The advantages for increasing investment activity by generous incentives (e.g. a broad scope of the tax base, refundability in case of losses) represent at the same time the more costly aspects of this measure. However, similar to accelerated depreciation schemes, these incentives materialise for new investments only and apply to certain categories of expenditure only (Thomson 2017). In order to further limit government expenditure, it is feasible to cap investment allowances beyond specific thresholds to ensure that incentives are only available to small or particularly sustainable firms (Clark and Sichel 1993; Hanappi, Millot, and Turban 2023).

One major limitation are the administrative and compliance costs caused by the complex and highly specific nature of these tax incentives (Hanappi, Millot, and Turban 2023). In addition, the delay in investment response, i.e. a time lag between carrying out the investment and receiving the tax benefit, could limit the effectiveness of tax incentives in promoting investment and innovation (Clark and Sichel 1993; Lome, Heggeseth, and Moen 2016). Finally, as these incentives are exclusively granted to the initially intended companies (Lerche 2022), separate incentives need to be developed and implemented for each asset or industry in which investments shall be encouraged, constituting an additional administrative burden for policy makers.

Depending on their design, R&D incentives can lower a firm’s effective tax burden significantly (see Section 3.1, Figure 7), potentially even below 15 percent. However, in the context of the global minimum tax, the jurisdictional blending, i.e. the aggregation of all group entities in a country, allows companies to compensate lower-taxed income (e.g. due to R&D credits, patent box regimes) with higher-taxed income (Ferreira Liotti et al. 2022; Perez-Navarro 2023). Therefore, the impact of the global minimum tax on R&D tax incentives is limited. Due to the jurisdictional blending approach, high-tax countries with tax incentives such as patent box regimes could even become relatively more attractive investment locations as companies have incentives to generate both ordinary-taxed and preferentially-taxed income there (Spengel, Schulz, and Winter 2023).