Tax Expenditure and the Treatment of Tax Incentives for Investment

Governments use tax expenditures to boost investment, innovation and employment. However, these schemes are largely opaque, costly and often ineffective in reaching their stated goals. They also frequently trigger unwanted side effects. In order to improve the performance of these tools, the authors present three concrete policy proposals: First, governments should increase transparency on tax benefits. G20 members should take the lead on this with frequent and comprehensive tax expenditure reports. Second, G20 governments should improve the design of tax incentives with the aim of minimizing the generation of windfall profits and negative spillover effects within and across (in particular, on poorer) countries. Third, governments should phase out tax expenditures that are environmentally harmful, including tax incentives for fossil fuels and other schemes that promote an unsustainable use of natural resources.


Challenge
Tax expenditures are costly and used widely as public policy instruments by governments worldwide. In the United States, tax expenditures are estimated to have cost the federal government more than 1.5 trillion USD in 2017 ( US Treasury, 2018). In the UK, tax expenditures amounted to 117 billion GBP in 2015 -not accounting for roughly 1 billion GBP in "minor reliefs" and 218 reliefs for which figures are not available; and in Australia the largest 25 reported tax expenditures added up to more than 131 billion AUD in 2015 (Houlder, 2016 andAustralian Treasury, 2016).
Strikingly though, transparency on the magnitude of existing tax expenditures is limited, and new tax benefits are being introduced regularly without adequate scrutiny. For instance, 10 out of 43 OECD/G20 countries do not report on tax expenditures at all and the official tax expenditure reports of the remaining countries are based on highly heterogeneous standards both with respect to the quality and scope of the data they provide (Neubig and Redonda, 2017). The picture does not improve when looking at developing economies as only a few African countries publish figures on tax expenditures, and in the East Asia and Pacific region, only the Philippines and Papua New Guinea compute and report tax expenditure estimates (AfDB et al., 2017 andWorld Bank, 2015).
As a result of the general lack of transparency, these schemes are not (as a rule) subject to sound cost-benefit analyses. Indeed, while studies assessing the effectiveness and efficiency in reaching the stated goals of tax expenditures are crucial, they remain rare.
Tax incentives for investment are a case in point. Most governments grant a myriad of tax benefits to stimulate investment. These include tax expenditures such as tax holidays, patent boxes, and special fiscal regimes offered in special economic zones (SEZs) to attract foreign direct investment (FDI). Nonetheless, these schemes usually have little impact on investment or growth, and operate instead as tax competition instruments that erode other economies' tax bases. This is a crucial issue not only for G20 economies, but also when it comes to domestic revenue mobilisation in developing economies. Clearly, more technical cooperation in this field is needed.
Moreover, several tax expenditures are environmentally harmful. Tax exemptions for the production and consumption of fossil fuels provide a concrete example. Many further schemes incentivize an inefficient and unsustainable use of natural resources. Yet, the negative environmental externalities triggered by such tax expenditures are very often neglected and, hence they are hardly ever internalized to assess the social cost of these schemes.
Against this background, and as recommended by the T20 last year, tax expenditure is an area on which the G20 can and should act urgently (Brosio et al. 2017).
Proposal 1 G20 member countries should report systematically and comprehensively on the fiscal cost of tax expenditures Rationale 1.
The tax expenditures debate should be about the pros and cons of a particular tool and not about the goals. 2.
An accurate measurement of tax expenditures is necessary to assess their effectiveness and efficiency.

3.
Comprehensive data on the scale and impact of tax expenditures is critical for international cooperation at the G20 level (and beyond).

4.
The G20 should step up technical cooperation to third countries with regard to measuring tax expenditures, strengthening domestic skills and facilitating the implementation of joint standards.

Discussion
Improving reporting on tax expenditure by systematically estimating and publishing their fiscal cost is crucial for governments to better scrutinize the effectiveness and efficiency of these measures. It is also critical to enhance transparency and accountability.
In 2017, the T20 already called for more and better technical cooperation among G20 governments, including the "compilation and publication of reliable tax expenditure data" as well as the standardization of the reporting of tax expenditures (Brosio et al., 2017). Unfortunately, progress in the field has been rare and -if anything -was mainly based on unilateral initiatives rather than on a coordinated strategy among G20 governments. We hereby propose a two-stage strategy to accelerate momentum. First, based on national tax laws and methodologies, governments should systematically estimate and report the revenue foregone through tax expenditures -ideally disaggregated by e.g, sector, tax base, policy objective, and as a percentage of GDP as well as of total tax revenue. In that context, pulling together existing official tax expenditure reports in a crosscountry database is a low hanging fruit that would make an important contribution towards transparency. There are, at least, three initiatives by international governmental organizations moving in this direction. The OECD/IEA Fossil Fuel Support Database 1 and the OECD Database on Tax Support for R&D and Innovation, 2 both provide crosscountry data for particular policy fields, disaggregated by direct government support and tax expenditure. The Inter-American Center of Tax Administrations publishes a regional Database on Tax Expenditure (DBTE). 3 Following the recommendations included in the CIAT Handbook of Best Practices on Tax Expenditure Measurements, the DBTE lists a total _________________________ number of just under 4500 tax expenditures included in the latest report of 16 Latin American economies (CIAT, 2011). Where available, the DBTE provides information regarding tax expenditure as a percentage of GDP for the 2005-2016 period (on average across countries, 3.5% in 2016). Schemes are classified by i) type of tax, ii) type of tax expenditure, and iii) sectors (Longinotti, 2017). In addition, this first stage should also include the design of a simple and standardized ("checklist" style) template to be used by those countries that currently do not report on tax expenditures at all. A step in this direction has recently been taken by UN and CIAT (2018). The joint publication -which seeks to provide tax policy makers and administrations with a methodology allowing them to assess the net benefits of tax incentive programs -includes a checklist for drafting tax incentives legislation enumerating several issues that should be considered and addressed to maximise clarity of scope and administration so that consistency between tax incentives and the underlying tax policy is ensured.
Second, governments should harmonize their methodologies to estimate the fiscal cost of tax expenditures. Measuring tax expenditures is a time and resource intensive task, which comes with significant methodological challenges, in particular regarding the choice of appropriate benchmarks. This said, efforts to standardize both the methodologies to estimate tax expenditures as well as their reporting will increase international comparability and enable efforts to define best practices. In that context, it would be helpful to run simulations to assess how different methods affect the reporting of tax expenditures in the same country. This would illustrate which methods lead to low or high estimates. It would also establish a range where the "true" value of tax expenditures is probably located.
Proposal 2 G20 member countries should improve the design of tax incentives for investment, avoid their use as tax competition instruments, and ensure policy space to appropriately adjust tax expenditures Rationale 1.
Tax incentives for investment should be designed to effectively impact real activity and avoid windfall gains for beneficiary businesses that reduce their efficiency.

2.
These tax incentives may create negative spillover effects and erode other economies' tax bases. The resulting race to the bottom is particularly worrisome for poorer countries and misaligned with the G20 objective to generate sustained, inclusive and sustainable growth. 3.
It is crucial to ensure governments have policy space to adopt appropriate reforms to improve tax expenditures, and that investment treaties are not used to stall or prevent policy shifts in this direction.

Discussion
Governments worldwide implement myriad tax incentives for myriad reasons. Often, the incentives are adopted with the stated objectives of attracting FDI. In addition to these "locational" schemes aiming to influence where investment occurs, tax incentives also seek to influence behaviour, such as whether investment takes place (e.g., whether a company decides to invest in R&D or other activities as opposed to holding its cash), and toward what activities the investment is directed (e.g., in fossilfuel based or renewable energy projects).
In terms of the tools used, advanced economies usually grant benefits channelled Trade, Investment and Tax Cooperation 7 through corporate income taxes (CITs), e.g. CIT credits and tax incentives for R&D. Developing countries instead, prioritize reduced tax rates as well as tax holidays for CIT, Value Added Tax (VAT) and excise taxes including, e.g. SEZs (IMF et al., 2015). Latin American countries mostly use CIT, import tariffs and VAT reductions/ exemptions to promote investment (Agostini and Jorrat, 2017).
There is conclusive empirical evidence regarding the positive effects of investment on employment, productivity and, ultimately, on inclusive economic growth (UNCTAD, 2014). Thus, sound strategies to boost investment should be a priority for governments worldwide. This said, benefits of investment are not automatic, and government policies (and the resources to implement these policies) are crucial for ensuring that positive outcomes are maximized, and harm minimized or avoided. In this context, the effective and efficient design of tax incentives for investment is critical. Their track record so far is misaligned with that objective.
Tax incentives for investment are usually poorly designed and ineffective. Hence, their impact on investment is often negligible and they are likely to trigger costly windfall gains for businesses. Innovation or patent boxes are a case in point. Whereas they may have a considerable impact on attracting patents, their effect on real activity is minimal as they create incentives for multinational enterprises (MNEs) to shift the location of their patents rather than modifying their real investment decisions (Alstadsaeter et al., 2015). SEZs provide another example. Despite a few success stories -mainly in Asian economies such as China, South Korea and Taiwan -empirical evidence shows a significant lack of effectiveness as SEZs often have a negligible impact on economic growth, employment creation, exports, and attraction of FDI. For instance, using micro data to evaluate Export Free Zones in Costa Rica, El Salvador and Dominican Republic, Artana (2015) finds that tax incentives "are redundant, create incentives to artificially readjust projects to keep receiving those benefits and favour tax avoidance through strategic tax planning taking advantage of subsidiaries located in eligible zones".
Where tax incentives for investment have an impact, they often also have a sizable price tag, thus limiting their efficiency. A 2017 study by the World Bank that uses firmlevel data for the Dominican Republic finds a positive effect of SEZs on employment creation that, however, comes at the expense of a significant fiscal cost, which in turn hinders the government's capacity to finance other investment and social services (World Bank, 2017). As it was discussed by a 2011 report for the G20's Development Working Group, "striking the right balance between an attractive tax regime for domestic and foreign investment, by using tax incentives for example, and securing the necessary revenues for public spending, is a key policy dilemma." (IMF, OECD, UN and World Bank, 2011).
The efficiency of tax incentives for investments is further limited through negative spillover effects on other countries. As discussed more in detail in this year's T20 Policy Brief on "Tax Competition", this is a crucial issue as corporate income tax competition has been mounting in recent years. There exists, for example, a high level of political and economic uncertainty regarding the final outcome once Brexit has taken place. Uncertainty also regards fiscal policy and taxation as the UK has announced a reduction of its CIT rate from the current 19% to 17% for the year starting 1 April 2020. Likewise, many commentators have forecasted a new race to the bottom after the US government passed the Tax Cuts and Jobs Act. Very likely, such a tax competition game will not only materialize in a reduction of statutory CIT rates but also in significantly lower effective tax rates, e.g. through the implementation of tax incentives for investment.
G20 governments should avoid using tax incentives for investment to exacerbate an already worrisome scenario in terms of international tax competition. Although important steps to mitigate these spillovers are already being taken -in particular through the OECD's Forum on Harmful Tax Practices (FHTP) to identify those schemes that "…facilitate base erosion and profit shifting, and therefore have the potential to unfairly impact the tax base of other jurisdictions" -negative spillover effects across countries continue to abound.
Moreover, whereas the definition of "harmful" used by the FHTP exclusively refers to external effects on other countries, tax incentives for investment are also likely to trigger efficiency issues and undesidered negative externalities within countries, e.g. horizontal inequities among different type of businesses.
An assessment of potential domestic spillover effects in terms of e.g. employment creation or productivity gains in non-eligible sectors is critical for a comprehensive review of tax incentives.
Against this background, G20 governments should improve the design of tax incentives for investment with the aim of minimizing the generation of windfall profits and negative spillover effects within and across (in particular, on poorer) countries. A peer review process similar to that of the FHTP based on a standardized methodology to estimate and report the fiscal cost of tax incentives (as discussed in Proposal 1) would contribute to a more responsible use of preferential tax regimes worldwide.
In that context, it is important to note that international investment treaties may limit, or be used to challenge, government efforts to reduce or eliminate tax incentives (Johnson, 2016). Even if an incentive program was not essential to an investor's decision regarding whether, where, or in what to invest, investment treaties may be interpreted as protecting that investor's ability to rely on the continued existence of such benefit. Consequently, governments may have to compensate foreign investors for changes to incentives schemes and offerings. While a few modern treaties such as the EU-Canada Comprehensive Economic and Trade Agreement (CETA) try to expressly address this issue, the vast majority do not. Against this background, G20 members should anticipate and limit claims resulting from tax expenditure reforms by, for example, emphasing the legitimacy of government efforts to evaluate and modify such programs, and the importance of policy space to take such actions.
Proposal 3 G20 member countries should reinforce efforts to phase out tax expenditures that are environmentally harmful, including tax incentives for fossil fuels and other schemes that promote an unsustainable use of natural resources.

1.
Many tax expenditures have direct and indirect negative effects on the environment and climate change. Identifying and minimizing the use of environmentally harmful tax incentives should be a priority for governments worldwide.

2.
The G20 has commited to "rationalise and phase out, over the medium term, inefficient fossil fuel subsidies that encourage wasteful consumption, recognising the need to support the poor" (G20, 2017). Tax expenditures related to fossil fuels are at the core of this commitment. 3.
The G20 aims for sustained, inclusive and sustainable growth. This objective cannot be achieved with an inefficient or unsustainable use of natural resources, which trigger global unbalances and negative externalities.

Discussion
Tax expenditures have a wide range of environmental implications. Fossil fuel subsidies are the most evident case in point.
The Inventory of Support Measures for Fossil Fuels provides an overview of the support that governments provide to the production and consumption of fossil fuels. 4 Roughly 60% of these measures are tax expenditures, including, e.g. reduced excise rates on aviation fuel in Australia, a special tax regime for goods used in the exploration and production of fossil fuels in Brazil, and an energy tax refund for diesel used in agriculture and forestry in Germany.
Despite this inventory being the most comprehensive source of cross-country information on the support for fossil fuels, the lack of transparency is -once again -a crucial problem. As stated in the 2015 Companion to the Inventory, "a limiting factor in respect of tax expenditures relating to fossil fuels is the extent to which countries release such estimates already" (OECD, 2015).

8
Tax expenditures relathaed to fossil fuels have been moving up agendas worldwide. Last year, the T20 explicitly called for G20 member economies "to reinforce their efforts in implementing the G20 commitment to phase out inefficient fossil fuel subsidies, including tax exemptions" (Brosio et al., 2017). World leaders adopted a similar recommendation in the Sustainable Development Goals (SDGs). 5 Strikingly though, concrete measures are rare, and progress remains slow.
Moreover, tax incentives linked to the consumption and production of fossil fuels are not the only environmentally harmful schemes. G20 governments should remove tax benefits incentivizing an unsustainable use of natural resources -including, e.g. deforestation -as they trigger undesired effects on resources' allocation and the environment, among others.
Finally, many tax expenditures have less straightforward negative effects on the environment, which are, however, non-negligible and should thus be internalized. The mortgage interest deduction (MID), a scheme allowing mortgage interest payments to be deducted from personal taxable income to foster homeownership, is a case in point. While its stated objective is an increase in homeownership, it offers an incentive to acquire a larger and more expensive house, which in turn increases its environmental footprint. As highlighted by the Tax Foundation, for instance, the US MID creates "negative externalities (i.e. pollution or greenhouse gas emissions) that likely exceed any positive external social benefit from marginally larger homes" (Prante, 2013). MID schemes are implemented in several countries including the US (where its fiscal cost is estimated to amount to more than 60 billion USD, hence being one of the largest federal tax expenditures in the country -see JCT, 2017), the Netherlands and Switzerland. In addition to its negative environmental impacts, empirical evidence indicates that the MID is also ineffective in promoting homeownership and highly regressive (see, e.g. Hilber and Turner, 2014).
The tax treatment of company cars and commuting expenses provides another example. As stated by the OECD, most member countries treat only 50% of the personal benefit to employees from company cars as taxable, which often creates incentives for employees to use company cars for personal use, and to drive longer distances than they might do otherwise.6 This triggers a negative impact on the environment, not only with regard to fossil fuel use, but also with regard to other pollutants and urban sprawl (see e.g. Heuermann et al., 2017). G20 governments should take concrete steps to comprehensively assess tax incentives including these kinds of negative externalities on the environment. Phasing out those schemes that have a significant negative effect on the environment would make a substantial contribution to shifting the economy to a low-carbon path that is aligned with the goal of fiscal sustainability and inclusive growth.

_________________________
5 SDG 12 includes an explicit reference to fossil fuel subsidies through taxation: Target 12.C: "Rationalize inefficient fossil-fuel subsidies that encourage wasteful consumption by removing market distortions, in accordance with national circumstances, including by restructuring taxation and phasing out those harmful subsidies, where they exist, to reflect their environmental impacts…".