100756 Options for Low Income Countries' Effective and Efficient Use of Tax Incentives for Investment             A REPORT TO THE G-20 DEVELOPMENT WORKING GROUP BY THE IMF, OECD, UN AND WORLD BANK Options for Low Income Countries’ Effective and Efficient Use of Tax Incentives for Investment A REPORT TO THE G-20 DEVELOPMENT WORKING GROUP BY THE IMF, OECD, UN AND WORLD BANK This report was prepared at the request of the G20 Development Working Group by the staffs of the International Monetary Fund, the Organisation for Economic Co-operation and Development, the United Nations and the World Bank. I t h a s benefitted from consultation with other organisations working in the tax area, officials of developing countries, Civil Society Organisations, and business representatives. The report is prepared under the responsibility of the Secretariats and Staff of the four organisations. It reflects a broad consensus among these staff, but should not necessarily be regarded as the officially-endorsed views of those organisations or their member states. The report was presented as requested to the G20 DWG in September, 2015, and to the Executive Board of the IMF for information, in October, 2015. OPTIONS FOR LOW INCOME COUNTRIES' EFFECTIVE AND October 15, 2015 EFFICIENT USE OF TAX INCENTIVES FOR INVESTMENT EXECUTIVE SUMMARY Experience shows that there is often ample room for more effective and efficient use of investment tax incentives in low-income countries. Tax incentives generally rank low in investment climate surveys in low-income countries, and there are many examples in which they are reported to be redundant—that is, investment would have been undertaken even without them. And their fiscal cost can be high, reducing opportunities for much-needed public spending on infrastructure, public services or social support, or requiring higher taxes on other activities. Effective and efficient use of tax incentives requires that they be carefully designed. Many low-income countries use costly tax holidays and income tax exemptions to attract investment, while investment tax credits and accelerated depreciation yield more investment per dollar spent. Tax incentives targeted at sectors producing for domestic markets or extractive industries generally have little impact, while those geared toward export-oriented sectors and mobile capital appear to be relatively effective—but the former need to be tempered by considerations of WTO consistency and both can be instances of mutually damaging tax competition. Enabling conditions—good infrastructure, macroeconomic stability, rule of law, etc.—are important for effectiveness. Good governance of incentives is critical for their effectiveness and efficiency. Transparency is necessary to facilitate accountability and reduce opportunities for rent seeking and corruption. Tax incentives should therefore be subject to legislative process, consolidated under the tax law, and their fiscal costs reviewed annually as part of a tax-expenditure review. The approval process of tax incentives may involve several stakeholders, but is ultimately best consolidated under the authority of the Minister of Finance and enforced and monitored by the tax administration. To the extent possible, the granting of tax incentives should be based on rules rather than discretion. Despite political obstacles, several countries have successfully reformed their tax incentive regimes. The proliferation of incentives is largely a manifestation of international tax competition—which regional coordination can help mitigate, although this requires political commitment and an effective supranational enforcement mechanism—which is often lacking. Common reporting standards and data collection can be an important first step toward coordination and enhanced transparency. More systematic evaluations are needed to facilitate informed decision making. In most low-income countries, the effectiveness and efficiency of tax incentives cannot be assessed due to lack of data and the absence of analytical tools and skills. The background document to this report offers guidance on how to develop the data and tools required for systematic analysis. Progress requires concerted action by several stakeholders to ensure evidence-based, transparent decision making. TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS CONTENTS INTRODUCTION __________________________________________________________________________________ 6 EFFECTIVENESS AND EFFICIENCY ________________________________________________________________ 8 A. Prevalence _______________________________________________________________________________________ 8 B. ‘Effective use’ __________________________________________________________________________________ 11 C. ‘Efficient use’ __________________________________________________________________________________ 16 GUIDANCE IN THE USE OF TAX INCENTIVES__________________________________________________ 19 A. Design _________________________________________________________________________________________ 19 B. Governance ____________________________________________________________________________________ 23 C. Reform ________________________________________________________________________________________ 28 INTERNATIONAL COORDINATION ____________________________________________________________ 29 CONCLUSIONS __________________________________________________________________________________ 32 BOXES 1. Elements of a Cost-Benefit Assessment of Tax Incentives _____________________________________ 10 2. Evidence on Productivity Spillovers from FDI __________________________________________________ 15 3. Transparency __________________________________________________________________________________ 24 FIGURES 1. Prevalence of Income Tax Incentives around the World _________________________________________ 9 2. Importance of Tax Incentive Packages Relative to Other Factors in Investor Location Surveys 12 3. Investment Incentives and FDI in Central and West Africa ____________________________________ 14 4. Governance of tax incentives by income group _______________________________________________ 25 4 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS TABLE 1. Redundancy of Tax Incentives Based on Investor Surveys _____________________________________ 12 APPENDIX I. Consultations___________________________________________________________________________________ 35 REFERENCES ____________________________________________________________________________________ 37 5 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS INTRODUCTION This paper responds to a request of the G20 Development Working Group for an exploration of options for low-income countries’ effective and efficient use of tax incentives for investment.1 To that end, it develops principles for the design and governance of tax incentives and provides guidance on good practices in these areas. Since much of the pressure to offer incentives stems from an awareness of those offered by other countries, the paper also discusses options for international coordination to address the risk of mutually damaging spillovers from such tax competition. Finally, a separate background document develops practical tools and models that can help assess the costs and benefits of tax incentives, which is essential for informed decision making. The aim is thus to assist low-income countries (LICs)2 in reviewing and reforming their tax incentives, so as to better align them with their developmental objectives. Investment tax incentives should be viewed as part of broader (tax) policy design. A good revenue system adopts taxes that are simple, fair and efficient. Tax incentives risk compromising these principles to the extent that they complicate the tax system, create horizontal inequities, and distort production efficiency; and they may forgo revenue that could have been spent more productively or needs to be replaced in other and more damaging ways. At the same time, tax policy may be able to play a purposive role in improving on market outcomes that are inefficient or unfair. The economic rationale for tax incentives must thus be evaluated in terms of their ability to achieve clear goals in ways that are both effective and efficient, relative to alternative policies, both tax and non-tax, that could achieve the same objectives. Tax incentives have been a focus of work by international organizations (IOs) for many years.3 This report draws on the ample experiences and insights that IOs have gained from interactions with countries. IOs and many other observers4 have often found tax incentives to be ineffective, inefficient and associated with abuse and corruption. As a result, they have frequently advised countries to remove them or to improve their design, transparency and administration. Yet, this advice has often had limited effect. The common reluctance to scale back incentives—perhaps even, as will be seen, a tendency for them to proliferate—may reflect vested interests, political inertia, and tax competition with other countries. It might also be that observers have underestimated the 1 Other countries of course also make extensive use of tax incentives, and much of the analysis below is also relevant to them. 2 Definitions of LICs differ, but usually refer to countries below a certain per capita GNP. The group may consist of between 30 and 60 countries, depending on the threshold. The analysis in this paper is relevant for a large group of developing countries and a broad definition may thus apply. 3 Studies by IOs include Lent (1967), UNCTAD (1996, 2000), Clark (2000), OECD (2001, 2006), Tanzi and Zee (2000), Zee, Stotsky and Ley (2002), Easson and Zolt (2004), Keen and Mansour (2010), Klemm (2010), James (2013; 2014) and Zolt (2014). Tax incentives are also frequently discussed and evaluated in technical assistance provided by IOs. 4 Some examples are Mintz (1990), Boadway and Shah (1992), Shah (1995), Bird (2000), Eason (2001; 2004), Bolnick (2004), Gugl and Zodrow (2006) and Bird and Zolt (2008). 6 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS benefits that incentives have generated. Drawing on recent insights from interactions with countries and an extensive range of academic and other studies, this paper aims to take a fresh look at incentive policies in LICs in order to develop practical advice for improvement, supported by simple tools for making progress.5 The analysis in this paper is necessarily selective. In particular, the focus is on:  Tax incentives, as defined in the next section—thus ignoring non-tax incentives, such as grants, in-kind benefits or loan guarantees: such other measures can in principle mimic the effects of tax incentives, but are usually designed differently and subject to different governance procedures.  …for investment—excluding tax incentives aimed at other objectives, such as supporting charitable giving or owner occupation or reducing pollution.6 The focus, moreover, is on incentives in the tax treatment of business income; governments may also seek to attract investment by special treatment in relation to Value Added Tax (VAT) or tariffs, for instance, but these are touched on only briefly—both for reasons of space and because income taxation and the associated policy concerns have been more prominent in the debate on tax incentives.  …that are implemented at the national level—those implemented at the sub-national level raise other, broader fiscal federalism issues, which go beyond the scope of this paper.7 This paper relates to other global initiatives aimed at strengthening domestic revenue mobilization in LICs. For instance, there are several initiatives to strengthen tax design and improve capacity in tax administration in LICs to enhance their ability to mobilize domestic resources, including through extensive and long-standing technical assistance by the IMF, World Bank and others. Moreover, the G20/OECD work on base erosion and profit shifting (BEPS) aims to support the implementation of BEPS outcomes to LICs, as suited to their circumstances, by providing toolkits. During consultations with developing countries on BEPS, a common concern expressed by LICs was the extent to which tax incentives erode their tax bases. This concern reflects an important tension, which was also emphasized by the IOs in their joint report for the DWG in 2011: “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 underlying report explores this dilemma in greater detail. 5 All case studies used in the paper aim to illustrate key problems, practices and achievements and are not intended to provide country-specific recommendations. The choice of our case material is inevitably selective and there could have been equally valid cases in other countries or periods. In selecting our examples, we used the following criteria: cases should i) be described in publicly available and easily accessible documentation; ii) obtain a broad regional coverage; iii) offer insight into either/both good and bad practices; (iv) preferably relate to experiences in LICs, although experience from other countries is used where this provides concrete and useful examples for LICs. 6 While tax incentives can be targeted to ‘green’ investment, this paper does not address this dimension specifically. 7 Note that interactions between national and sub-national tax incentives covering the same activities or assets risk producing unintended and potentially perverse results, such as negative tax burdens for certain industries. 7 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS EFFECTIVENESS AND EFFICIENCY A. Prevalence In this paper, by a ‘tax incentive’ is meant any special tax provisions granted to qualified investment projects or firms that provides favorable deviation from the general tax code. They can take several forms, such as tax holidays (complete exemption from tax for a limited duration), preferential tax rates in certain regions, sectors or for certain asset types, or targeted allowances (tax deductions or tax credits) for certain investment expenditures. Tax incentives have been used to pursue a variety of objectives. The primary motivation is usually to stimulate investment and—especially in LICs—attract foreign direct investment (FDI). FDI inflows, for instance, are believed to not only bring capital and (high-wage) jobs to a country, but can also spur competition and increase the efficiency of domestic markets more widely, thus contributing to a country’s overall economic development. Empirical growth regressions indeed generally find positive correlations between inward FDI and economic growth, although conclusions about causality remain contentious (see e.g. Adams, 2009, for a review of evidence on Africa). Tax incentive policies also often aim to promote specific economic sectors or types of activities as part of an industrial development strategy or to address regional development needs. They are commonplace around the world (Figure 1). The use of tax incentives is widespread, and extends well beyond LICs. Countries differ with respect to the type of incentives used, with high- income countries relying more on investment tax credits and favorable tax treatment of research and development (R&D), low-income countries relatively more often offering tax holidays and reduced tax rates, and middle-income countries most often having preferential tax zones (in which income can be tax exempt and other favorable treatments may apply). Over the last decades, tax incentives have become more widespread in LICs. For instance, in 1980, less than 40 percent of the LICs in sub-Saharan Africa offered tax holidays while free zones were non-existent. By 2005, more than 80 percent offered tax holidays and 50 percent had adopted free zones (Keen and Mansour, 2010). The number of countries in sub-Saharan Africa granting tax holidays and establishing free zones has grown further since (James, 2014). There is also evidence, however, that the average length of tax holidays has declined somewhat in various regions of the world (Abbas and Klemm, 2013). With the global declining trend in corporate income tax (CIT) rates, including in LICs, it might indeed be that the benefit for investors of receiving tax incentives has somewhat diminished. Broadly, a tax incentive serves a useful social purpose if the social benefits it generates exceed the associated social costs. The background document to this report offers a simple conceptual framework that lends practical substance to this general principle by identifying the key 8 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS components that affect the social benefits and costs of tax incentives: these are set out in Box 1.8 The next subsections look in turn at the benefit side of the calculation—the ‘effectiveness’ of incentives in achieving their aims—and then the cost side—their ‘efficiency’.9 Figure 1. Prevalence of Income Tax Incentives around the World /1 /1 Figure shows the percent of countries in each of four income groups that have the indicated incentive. The sample size per income group is denoted between brackets. Source: Calculations based on James (2014) 8 Sometimes the term ‘wasteful tax incentives’ is used. This concept is not always well-defined, however. This paper adopts the framework of cost-benefit analysis to identify whether tax incentives are desirable or not. 9 In principle, this should be assessed taking into account also other ways of achieving the aims of policy: the benefits associated with a regional tax incentive might be achieved at even lower social cost, for instance, by a regional spending program. The point is of considerable practical importance, but since the alternative instruments are highly context-specific, cannot be pursued in detail here. 9 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Box 1. Elements of a Cost-Benefit Assessment of Tax Incentives Investment tax incentives ultimately aim to contribute to a country’s development and improved living conditions for its citizens. As elaborated on in the background document, the following elements are critical for the social benefits:  Size of the net investment effect—the rise in investment should be corrected for redundancy effects (investments that would have occurred without the incentive) and displacement effects (the reduction in any other investments) to infer the net incremental increase in capital due to the incentive.  Net impact of higher investment on jobs and wages. New jobs can yield significant social gains if they reduce unemployment. However, if new jobs displace existing jobs, the social benefits depend on the productivity (and wage) differential between the new and old jobs.  Productivity spillovers. To the extent that new investment boosts productivity elsewhere in the domestic economy, such as in supplying or competing firms (often seen as a particular benefit from inward FDI), this magnifies social benefits by raising income levels more widely. The social costs of tax incentives depend on the following factors:  Net public revenue losses—public revenue falls if tax incentives are redundant or create leakage and abuse. But additional net investment and jobs can recover some of the revenue loss.  Administrative and compliance costs, which can rise due to tax incentives, especially if they are complex or create opportunities for rent seeking and corruption.  Scarcity of public funds. Often overlooked is that $1 of tax revenue has a higher social value than $1 of private income, because it is the greater value of the public expenditure it finances that justifies transferring resources from public to private sectors through distortionary taxes. To compare changes in private income and tax revenue, the latter thus need to be weighted by the ‘marginal cost of public funds’, which will be greater than unity (as discussed further below).  Distorted resource allocation. Discrimination in favor of some and against other investment implies that taxes, rather than productivity differences, determine resource allocation. This distortion reduces average productivity and lowers income per capita. 10 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS B. ‘Effective use’ Effective use is taken here to mean that tax incentives realize their stated objective. This can be assessed irrespective of associated costs, which will be explored later. Raising investment or FDI is usually a necessary but not a sufficient condition for ‘effectiveness’, as the higher investment should be of the kind envisaged to yield the desired social benefits in broader welfare terms. Empirical evidence finds that taxes matter for investment, although most likely less so in developing countries. Empirical studies on the relationship between effective tax burdens and FDI generally conclude that host country taxation significantly affects investment (De Mooij and Ederveen, 2008).10 Most of this evidence, however, refers to advanced economies. Recent studies report similar results for developing countries, although the effects tend to be somewhat smaller on average (James and Van Parys, 2009; Abbas and Klemm, 2013). One reason might be that many developing countries do not offer attractive general investment conditions for most multinational companies, due to for instance poor infrastructure, macroeconomic instability, unclear property rights, and weak governance or judicial systems. In these circumstances, tax incentives do not effectively counterbalance such poor conditions and are largely ineffective (Kinda, 2014). At the same time, however, tax incentives might be one of the few (albeit second-best) instruments for LICs to offset disadvantaged circumstances, address regional disparities and mitigate market failures, such as lack of financial access. Investment surveys confirm that tax incentives usually do not top the list of investment factors in developing countries. In 2010, the United Nations Industrial Development Organization conducted a business survey of 7,000 companies in 19 sub-Saharan African countries active in agriculture, mining, manufacturing, utilities, construction, and services sectors. Investors were asked to rank the importance of twelve location factors and to assess how they might have changed, improved and worsened, in the last three years (Figure 2). The results suggest that tax incentives packages ranked 11th out of 12 in importance; and this importance fell over time. For comparison: transparency of the legal framework ranked 5th in investors’ concerns and grew in importance. Investors thus seem to care much more about deficient legislation and onerous regulations than about the availability of tax incentives (UNIDO, 2011). Tax incentives are often found to be redundant in attracting investment in developing countries; that is, the same investments would have been undertaken even if no incentives had been provided. Table 1 shows redundancy ratios, based on investor surveys in various countries, measured by the percentage of investors who claim that they would have invested even without tax incentives. Redundancy levels thus obtained—subject to well-known caveats, such as a discrepancy between answers and actual behaviors under a counterfactual scenario—are high in most countries. For example, redundancy rates exceed 70 percent in 10 out of the 14 surveys listed in Table 1. In 10 FDI comprises different components, such as new plants and equipment, plant expansions, and mergers and acquisitions (M&A). The evidence is that new plant location choices are more responsive to tax than are incremental investment increases; and M&A tends to respond to taxation differently from greenfield investment. 11 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Guinea, Rwanda, Tanzania and Uganda, more than 90 percent of the investments would, it seems, have been made even without the incentives. Figure 2. Relative Importance of Tax Incentive Packages in Investor Location Survey 2010 (Left: Relative Rank; Right: Change in Rank since 2007) 0 1 2 3 4 -0.3 0.0 0.3 Economic stability 3.82 0.31 Political stability Political stability 3.81 0.28 Local markets Costs of raw materials 3.65 0.24 Availability of skilled labour Local markets 3.63 0.15 Economic stability 0.12 Quality of life Transparency of legal framework 3.51 0.10 Transparency of legal framework Availability of skilled labour 3.50 0.09 Availability of local suppliers Labour costs 3.46 0.06 Bilateral agreements and treaties Quality of life 3.32 0.00 Export market Availability of local suppliers 3.22 -0.01 Labour costs Bilateral agreements and treaties 3.05 -0.22 Costs of raw materials Incentives package 3.05 -0.26 Incentives package Export market 2.94 Source: UNIDO (2011) Table 1. Redundancy of Tax Incentives Based on Investor Surveys /1 Burundi (2011) 77 Rwanda (2011) 98 El Salvador (2013) 37 Serbia (2009) 71 Guinea (2012) 92 Tanzania (2011) 91 Jordan (2009) 70 Tunisia (2012) 58 Kenya (2012) 61 Uganda (2011) 93 Nicaragua (2009) 15 or 51 /2 Vietnam (2004) 85 Mozambique (2009) 78 Thailand (1999) 81 /1 Percent of affirmative answers to the question if an incentive was redundant; /2 51 percent for non-exporting firms outside free zones. Source: James (2014) 12 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Effectiveness varies between countries and sectors. In some countries, tax incentives seem to have played an important role in attracting new investment and spurring economic growth. Famous examples include Korea and Singapore, where tax incentives—offered as part of a broader strategy to attract investment—seem to have encouraged rapid industrialization (Tanzi and Shome, 1992). However, in many instances tax incentives have resulted in little or no new investment, as indicated by the redundancy ratios in Table 1. Studies offer some insight into the factors determining effectiveness. For instance, FDI that is resource-seeking (to exploit the presence of natural resources), market-seeking (to penetrate a local market), or strategic asset-seeking (to exploit local know-how or technology) is generally found to be less responsive to tax than FDI that is efficiency- seeking (to exploit cost advantages in production for the world market).11 Indeed, tax incentives tend to have the greatest salience where investment is oriented toward exporting firms (Grubert and Mutti, 2004). China is often quoted as an example of effective (tax) incentive policies. During its transition period between the mid-1980s and mid-2000s, it experimented with a wide range of industrial policy instruments, including tax incentives for special economic zones, reduced tax rates for FDI, and tax holidays for strategic industries. FDI inflows accelerated during this period and the country became a top destination for many multinationals. In a panel of 29 regions between 1985 to 1995, Chen and Kwan (2000) find, for instance, that special economic zones systematically boosted FDI inflows. Examples of less effective tax incentives can be found in Africa. Figure 3 shows how FDI was influenced by changes in investment codes between 1994 and 2006 in countries of the CFA franc zone and the Economic Community of Central African States. The vertical lines in the figure denote the introduction of new investment codes, such as tax incentives and legal protections. Providing more generous tax incentives did not have any demonstrable effect on FDI (Van Parys and James, 2010). The effectiveness of incentives in attracting investment also depends on the international tax rules in place. Multinationals taxed on a “territorial” basis in their home country are able to retain the benefits of host country tax incentives, since there is no offsetting home country tax on the foreign source income. Multinationals might be subject to home country tax on foreign source income due to controlled foreign corporation (CFC) rules12 or tax upon repatriation under a ‘world- wide’ system—as used in e.g. the US, China and India. The tax incentive can then become ineffective, since the benefit will be taken away by increased tax payments in the multinational’s home country—although tax deferral until repatriation of income often effectively mitigates such an effect.13 11 This classification of types of FDI is due to Dunning (1993). 12 CFC rules vary, but are in essence provisions that bring immediately into tax passive income—the complement of active business income—arising abroad that has not paid at least some minimum amount of tax. For worldwide countries, CFC rules in principle provide some protection against tax avoidance through deferral; for territorial countries, they simply ensure that only active income is exempt in the residence country. 13 Moreover, some tax treaties provide for ‘tax sparing’, whereby the residence country of a corporation determines the foreign tax credit based on the taxes that would have been paid if the incentive had not existed. 13 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Figure 3. Investment Incentives and FDI in Central and West Africa BEN BFA CAF CIV 40 20 0 -20 CMR COG GAB MLI FDI perc GDP 40 20 0 -20 NER SEN TCD TGO 40 20 0 -20 1995 2000 2005 1995 2000 2005 1995 2000 2005 1995 2000 2005 year FDI perc GDP change inv climate Graphs by country_code Source: Van Parys and James (2010) The ability of multinationals to avoid host-country taxes may blunt the impact of tax incentives—so measures to restrict that ability may make tax incentives more effective. A multinational that can readily avoid host-country taxes (by, for instance, using interest deductions to shift profits to a low tax jurisdiction) may see little additional benefit from tax incentives. By the same token, actions that make such avoidance more difficult, from the G20/OECD BEPS project for example, would make tax incentives more attractive for them. Where tax incentives increase FDI, domestic investment may be displaced. Displacement reduces effectiveness in terms of the net impact of the incentive on the domestic capital stock. This happens, for instance, if FDI reflects a mere transfer of ownership, through mergers and acquisitions, or if domestic investment is ‘round tripped’ through a foreign entity to take advantage of the tax incentive. Displacement can also occur in labor markets, where jobs in new firms come at the expense of employment in other sectors of the economy. Evidence for 40 Latin American, Caribbean and African countries between 1985 and 2004 suggests that changes in the length of tax holidays systematically increased FDI inflows. These FDI inflows did not, however, increase total investment, nor did they increase economic growth. This suggests full displacement of domestic by foreign capital (Klemm and Van Parys, 2012). FDI inflows can yield various other social benefits, such as economic diversification gains, knowledge and technology spillovers, new management practices, reduced unemployment, and improved conditions in less-developed areas (Blomstrom and Kokko, 1998; OECD, 2002). FDI spillovers may impact other firms in the same sector (‘horizontal’ spillovers) and/or 14 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS supplying/purchasing firms (‘vertical’). The available evidence—mainly for emerging economies, rather than LICs—indicates strong empirical support of vertical spillovers, but weaker support for horizontal spillovers (Box 2). Box 2. Evidence on Productivity Spillovers from FDI Empirical studies of horizontal spillovers look at the systematic variation of productivity growth in an industry and its intensity of FDI. Early studies for Morocco, Russia and Venezuela find no support for such productivity spillovers in manufacturing industries; instead, and counterintuitively, they all report negative correlations (Haddad and Harrison, 1993; Aitken and Harrison, 1999; Yudaeva et al., 2003). Gorodnichenko et al. (2007) find that horizontal spillover effects are generally insignificant in an analysis for 17 countries in Eastern Europe and Central Asia. In a meta-analysis of 32 empirical studies on technology spillovers from FDI, Woodster and Diebel (2006) conclude that intra-sectoral FDI spillovers are non-existent in developing countries. Evidence for advanced economies is usually more supportive of horizontal spillovers. For instance, studies using data for the US and the UK typically report positive correlations between domestic plants’ productivity and FDI intensity (Xu, 2000; Keller and Yeaple, 2003; Haskel et al., 2007). Here, spillovers also tend to be more prevalent in high-technology sectors and when own R&D is undertaken, reflecting a greater ability to understand and assimilate new technologies (Griffith et al., 2004). Lack of absorptive capacity may explain why horizontal spillovers are less prevalent in developing countries. Studies on vertical spillovers usually explore backward effects of FDI to domestic suppliers, again by measuring productivity gains in the manufacturing sector. A study for Zambia, for instance, finds significant knowledge transfers from foreign to local firms (Bwalya, 2006). Similar positive spillover effects are found for Indonesia and Lithuania (Javorcik, 2004; Girma et al., 2007). For the 17 countries in Eastern Europe and Central Asia, Gorodnichenko et al. (2007) consistently report positive backward productivity spillovers. For strategic industries in China, Du et al. (2011) find support for backward and forward vertical FDI spillovers. 15 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS C. ‘Efficient use’ Efficient use in this paper means that objectives are achieved at low social costs. Such costs include revenue losses for government and other social costs, for example due to less efficient resource allocation. Redundancy matters for efficiency too, since it implies a loss of government revenue from projects that would have been undertaken also without tax incentives. Redundancy implies that the tax incentives are a mere cash transfer to the investor: a net social loss to the extent that the marginal cost of public funds exceeds unity (and an even greater loss in national terms if the investor is foreign). On the other hand, for projects that would not have been undertaken without the incentive, there is no direct revenue loss—so long as taxation of the incentivized activity is not entirely eliminated, there may in fact be a net revenue gain from those projects. To minimize the revenue cost of tax incentives, the goal would thus be to offer tax incentives only to those marginal investors who would not have invested otherwise. Indirect revenue costs arise from taxpayers abusing the tax incentive regime.14 For example, if tax incentives are only available to foreign investors, local firms may use foreign entities to route their local investments in order to qualify. Similarly, if tax benefits are available to only new firms, taxpayers may reincorporate or set up new corporations to be treated as a new taxpayer under the tax incentive regime. Other leakages occur where taxpayers use tax incentives to reduce the tax liability from non-qualified activities, for instance, by shifting taxable income to a related firm that qualifies for a tax holiday or that resides in a tax-free economic zone (McLure, 1999; Eason, 2004). Preventing such losses requires proper anti-abuse rules and strong administrative capacity to enforce them.15 In 2000, the government of India removed incentives for exporters, except those located in export processing zones or qualified as export-oriented units. Investment behavior hardly changed due to this reform. Indeed, firms that lost their incentives maintained the same level of investment as before, despite higher tax rates, similar to the control group that kept their incentives. However, reported profits did respond aggressively to the loss of incentives. In particular, reported pre-tax profits dropped by half on average in firms that lost their incentives, despite little change in sales. In contrast, pre-tax profits in firms that kept their incentives showed an increase. Hence, companies seem to have diverted profits from affiliates facing higher taxes to those exempt from tax due to the incentives (James, 2007). In 1987, Bolivia established a tax credit for exporters based on the growth of their export receipts over the previous year. This led to the emergence of what was called “tourist cows”: 14 See also the top 10 of most frequent abuses listed in Zolt (2014). 15 Anti-abuse rules can be either specific provisions in the tax law to prevent certain behaviors that are deemed abusive, or general provisions under which behaviors can be classified as abusive based on a broad characterization in the law. High-quality legal drafting and a robust administration are necessary to ensure effective anti-abuse provisions. 16 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS farmers generated tax credits by moving their herds back and forth across the border in order to maintain or increase the growth of their “export” receipts. The scheme was eliminated in 1991 (Rodrik, 1993). Additional resource costs arise for the government in administering tax incentives and for businesses in complying with the associated requirements. Of course, any tax comes along with such costs. However, they generally increase with the complexity of the assessment processes of tax incentives, and with the opportunities for rent seeking and corruption they might create (as discussed further in the next section). The additional administrative costs are a particular concern in LICs, where administrative capacity is often limited. Indeed, scarce resources might be diverted away from core aspects of a country’s tax administration, undermining other efforts to mobilize revenues. In Papua New Guinea (PNG), the taxation review committee recently evaluated its tax incentive policies and concluded that one significant concern is the additional complexity they create, placing greater demands on already limited tax administration resources. In particular, PNG’s experience with the research and development incentive and the infrastructure tax credit have highlighted the challenges of implementing or effectively monitoring these incentives in the absence of sufficient administrative or technical capacity (PNG Tax Committee, 2014). A first step to understanding the public revenue forgone as a consequence of tax incentives is to calculate the implied ‘tax expenditure’ (see background document). Investment tax incentives are only one form of tax expenditure, by which is meant a provision in the tax code that deviates from some benchmark tax system in a direction favorable to the taxpayer. A tax expenditure review quantifies the revenue forgone for each provision, including for investment tax incentives analyzed in this paper.16 Conceptual complexities arise when performing a tax-expenditure review, including in defining the relevant benchmark to which tax incentives are to be compared. Importantly, a tax expenditure review does not take into account either any effect of the incentive on investment or the leakage and abuse to which it can give rise to. The former may imply an overstatement of actual revenue cost, since elimination of the incentive might lead to a reduction in the tax base and hence to less additional revenue (if the rate would still be positive) than a ‘static’ calculation implies. The latter implies an understatement. As methodologies differ, international comparisons of tax expenditures are usually difficult. Tax expenditure analyses are increasingly being used by developing countries, with strong encouragement by and support from the IOs (see e.g. IMF, OECD, UN and World Bank, 2011), including through technical assistance. The International Budget Partnership’s Open Budget Survey indicates that tax expenditure reporting is undertaken by the following middle and low-income countries: Argentina, Chile, Columbia, Dominican Republic, Guatemala, India, Jamaica, Jordan, Kenya, Lebanon, Malaysia, Morocco, Nepal, Pakistan, Peru, Philippines, Russia, Serbia South Africa, 16 For assessing the revenue forgone from investment tax incentives, corporate micro simulation models can play an important role, as the background document explains further. 17 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Sri Lanka, Tanzania, Trinidad and Tobago. Some other countries carry out such analyses, but their score is low, indicating weak methodological frameworks. In the Philippines, tax incentives are provided to a large number of firms producing for the domestic market. The cost of fiscal incentives granted by the Board of Investments (which is only one of the authorities that can do this) in 2004 was estimated to exceed 1 percent of GDP (Reside, 2006). The background document discusses recent progress in the Philippines, with the Tax Incentives Management and Transparency Act. For purposes of policy evaluation, each dollar of public revenue forgone should be inflated by an indicator reflecting the scarcity of public funds: the ‘marginal cost of public funds’. Taxes generally reduce labor supply, saving and investment, thereby imposing an additional cost to society. The marginal cost of public funds (social cost of a dollar of tax revenue) in advanced economies is generally estimated around $1.20 to $1.30 due to these distortions. Estimates for developing countries are scarce, but constraints on the instruments for domestic revenue mobilization available to them likely mean that in LICS too, the marginal cost of public funds substantially exceeds unity. For instance, Auriol and Walters (2012) estimate it at an average of $1.20 for a group of 38 African countries. Intuitively, spending a marginal dollar on tax incentives by the government to attract FDI competes directly with other uses of funds, such as tax rate cuts across the board or increases in expenditure on infrastructure or education. High returns to such public investments in LICs due to a scarcity of public funds raise the opportunity cost of tax incentives by (on these estimates) an extra 20 cents per dollar revenue forgone, and thus call for a 20 percent higher payoff to justify them. Calculations of ‘dollar cost per job created’ are a popular metric for measuring the cost- effectiveness of tax incentives. Number of jobs can be measured either for all investors enjoying the incentive, or only for those that are ‘marginal’, in the sense that they would not have invested without the incentive. Dollar costs can either be based on total revenue forgone from the tax- expenditure review, or only those from the non-marginal investors. Clearly, calculations of this kind are simplistic and suffer from several pitfalls, implying they should be interpreted with caution. Still, however calculated, the dollar cost per job provides a quick ballpark figure that can inform policymakers on the relative cost-effectiveness of a particular incentive—which can also be compared with the costs of creating jobs by direct spending measures. Sometimes, the numbers calculated are striking. A 2008 Investment climate advisory study found that the government of Yemen spent about $6,000 per worker each year for 8,000 jobs that investment incentives helped create—more than six times the country’s per capita income. In Thailand, a 1999 study by the Foreign Investment Advisory Service found that investment incentives each year cost the government about 16 times the average annual wage of an industrial worker. In Tunisia, it was found that the cost of tax incentives for each job created was three and a half times the per-capita income. The government of India granted over US$200,000 per job in incentives. Investments by General Motors in Hungary cost US$ 300,000 per job created. (James, 2014; Kenneth, 2011) 18 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Inefficiencies also arise from distorted resource allocations. By definition, tax incentives place non-incentivized investments at a competitive disadvantage. The risk is that, in seeking to pick winners, there will be inefficiently high investment in incentivized activities and inefficiently low investment in others. Incentivized firms may also be able to attract workers from non-incentivized firms by offering higher wages, just because they enjoy an artificial competitive advantage. Diversion of labor and capital to the incentivized firms in response to discriminatory tax treatment will distort the allocation of resources and can hurt economic growth. The World Bank Group finds that a few politically connected firms have captured tax incentives in Egypt and Tunisia. This selective access caused a dual economy with large differences in profitability between insiders and outsiders, undermining a level playing field and reducing competition, significantly reducing job growth (Shiffbauer et al., 2015). In principle, efficiency may require taxing activities that are more mobile across countries less heavily than those less mobile…. This is in line with the standard principle of efficient tax design that items with less tax-sensitive bases should be taxed more heavily—because the tax then has a lesser impact on real decisions. Such an outcome would effectively be one in which incentives are offered to the most mobile activities—which in practice has clearly been a major concern for many LICs and others. Differential tax rates across sectors may also serve as a pragmatic device for imposing higher burdens where rents (that is, profits in excess of the minimum required by investors) are more substantial. …but this can bring its own difficulties, and be inferior to more cooperative outcomes. Applying different rates to different sectors, for instance, creates opportunities for profit shifting between the two (as elaborated on above). And since tax bases are less mobile from a collective perspective (that of a trading bloc, for instance) than between individual countries, low tax rates set in pursuit of national objectives may forego opportunities for efficient taxation from a collective perspective. Where there is reason to suppose substantial rents are being earned, targeted taxes may be the better way to approach them rather than by differential rates—hence the distinct resource rent taxes often found in, and recommended for, resource-rich countries. GUIDANCE IN THE USE OF TAX INCENTIVES As the analysis above makes clear, how tax incentives are designed and governed is critical to their effectiveness and efficiency. This section discusses principles and good practices in these areas, and elaborates on issues of reform of tax incentives. A. Design Tax incentive policies involve three core design issues:  Choice of tax instrument to incentivize investment;  Eligibility criteria used in the selection of qualified investments; 19 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS  Reporting and monitoring requirements during different stages of the tax incentive’s life cycle, as well as sunset and recapture provisions. Instrument choice Incentives can cover a wide range of taxes, including corporate income tax (CIT), VAT, tariffs, property taxes, personal income taxes and social contributions. Incentives in each of these areas require distinct economic analysis. For instance, a reduction in tariffs for capital goods is usually on solid economic ground in that it eliminates production distortions that create large welfare costs. On the other hand, VAT exemptions on investment may be entirely redundant, since full operation of the tax means that VAT charged on inputs does not ‘stick’ with the purchaser, but is fully recovered as a credit against VAT charged on sales;17 but VAT exemptions can become beneficial for firms if VAT implementation is problematic, for instance, due to imperfect VAT refund procedures. Corporate income tax incentives are probably the most contentious and widely criticized by observers. Much of the focus below will consequently be on them. Tax incentives that lower the cost of investment are often to be preferred over profit-based tax incentives:  Cost-based tax incentives involve specific allowances linked to investment expenses, such as accelerated depreciation schemes and special tax deductions and credits. They are targeted at lowering the cost of capital and so make a greater number of investment projects more profitable at the margin—that is, may generate investments that would not otherwise have been made.  Profit-based tax incentives generally reduce the tax rate applicable to taxable income; examples include tax holidays, preferential tax rates or income exemptions. One effect is thus to forego government revenue in order to make even more profitable investment projects that would be profitable, and hence undertaken, even without the incentive. The difference between the two types of instruments is critical. For instance, profit-based incentives will be less effective in encouraging investment compared to incentives that reduce the capital cost if profitability is low. When profits are earned due to the presence of location-specific factors, such as natural resources, agglomerations, or local markets, profit-based incentives tend to be associated with high redundancy rates and are again ineffective in raising investment. But international considerations are important here. Profitable investments that are highly mobile across national borders—because, for instance, rents are associated with intangible assets, such as patents or trademarks, that are easy to move between jurisdictions—may be sensitive to both cost-based and profit-based tax incentives. Tax holidays tend to favor readily mobile (‘footloose’) activities rather than long-term investment. By offering temporary tax relief for profitable firms, tax holidays benefit industries that 17 Indeed VAT exemption on investments (and intermediate purchases more generally) can be worse than redundant, because of its cascading effect through the production chain: see for instance Ebril et al. (2001). 20 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS start making profits soon in the holiday period. This introduces a bias towards short-term projects with low upfront investment costs, which may be those least likely to generate spillover effects on the wider economy (of the kinds described above). Such investment projects are known to “pack and go”, fleeing the host country as soon as the preferential treatment is removed. For industries with significant long-term capital needs, and for which spillover benefits may be greater, tax holidays could actually discourage investment: little otherwise taxable profit might be expected during the holiday period, and, to utilize depreciation allowances, a firm might be encouraged to postpone investment until after the holiday period in order to claim full deductions for its costs (see background document). Eligibility criteria Tax incentives need to be well-targeted and based on clear eligibility criteria. Targeting serves two related purposes: (i) identifying the types of investment that a government seeks to attract; and (ii) reducing the fiscal cost of incentives. The following criteria are commonly used—often in combination:  Special size. Tax incentives are sometimes restricted to new investments (or investors) that exceed some stipulated value of assets or those that create at least some stipulated number of new jobs. This of course has significant appeal, for instance where investments can be transformational for a country or region or where financing and technical constraints hold up investment. Limiting incentives to large investments can also reduce administrative costs for government. However, discrimination in favor of large foreign investments can also lead to manipulation, abuse and distortion. For instance, size conditions are relatively easy to meet on paper, but extremely difficult to monitor and verify ex-post. If an investor increases the size of their planned investment or the number of new jobs simply to obtain a tax privilege, this implies an inefficient use of resources, so that marginal productivity increases can be very low or even negative. Discrimination can also distort competition and restrict the growth of smaller domestic enterprises that do not enjoy incentives, even when they are more productive.18  Special sectors. Many countries grant preferential tax treatment to certain sectors of the economy, which policy makers consider as most desirable and which are most likely to be influenced by tax. Among the activities commonly preferred are tourism, “offshore” financial centers, film production and manufacturing activities, with the idea that they bring more socially valuable spillover effects. Incentives are also sometimes restricted to so-called “pioneer” industries, which can be defined in various ways, but are always viewed as of strategic value for 18 Many countries also offer special tax incentives for small and medium-sized (usually domestic) enterprises (perhaps to address their difficulties in raising external finance), for example through lower tax rates or special allowances. There is no clear evidence, however, that targeted tax relief for small firms is more cost-effective than general tax relief for businesses (IMF, 2012). On the contrary, special relief may hurt economic growth by creating a small-business trap, preventing small firms from growing larger to maintain their preferential tax treatment. 21 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS the national economy.19 There is always a question, however, whether serving special interests aligns with serving the general public interest. For instance, investments in other economic sectors are placed at a competitive disadvantage and may under develop, even though they are more productive.  Special regions/zones. Tax incentives are sometimes targeted to special regions in the form of ‘zones’, for example to address geospatial inequality. Economic zones (EZs)—which exist under various names and definitions20—have rapidly gained popularity over the last couple of decades. For instance, the International Labour Organization (ILO) estimated in 2007 that 3,500 EZs were operational in 130 countries, compared to only 176 in 46 countries 20 years earlier. Successful economic developments in Hong Kong SAR, Singapore, and Korea have often been attributed to EZs. In China, SEZs were used as test beds for the gradual liberalization of the economy and export-oriented industrialization. There is evidence that these Chinese SEZs, on average, boosted the growth rate of FDI by almost 7 percent per year (Wang, 2013) and regional economic growth by 12 percent per year (Alder et al., 2013). These successes in East Asia have inspired many other developing countries to adopt EZs of various kinds, including in sub-Saharan Africa and Asia. Experience on their effectiveness, however, is mixed at best. Zeng (2015) recently concluded, for example, that the experiences have been disappointing in most African countries, which he explains by poor institutions, weak governance and inadequate infrastructure (among others). In India too, there is evidence that the number of EZs per region has had no discernible impact on regional economic growth (Leong, 2013). In such circumstances there is evident risk of significant revenue loss through redundancy, and many observers also stress sizable risks from tax planning between the free-zone and affiliates outside the zone (similar to those shown in the India example above). “Performance data are elusive because the effects of zones are hard to disentangle from other economic forces. But anecdotal evidence suggests that they fall into three broad categories: a few runaway successes, a larger number that come out marginally positive in cost-benefit assessments, and a long tail of failed zones that either never got going, were poorly run, or where investors gladly took tax breaks without producing substantial employment or export earnings” (The Economist, April 4th 2015). 19 Investment incentives are also frequently targeted at export-oriented production. However, such incentives may be contrary to WTO rules—although there is a LIC exemption provided under the WTO Subsidies and Countervailing Measures Agreement, under which 33 LIC WTO members (as well as 12 accession LICs) are granted special relief. 20 Export processing zones (EPZs), for example, are enclaves where foreign companies engaged in the manufacturing of products for exports enjoy preferential (tax) treatment compared to the rest of the economy. Special economic zones (SEZs) offer locational flexibility and have a wider application than EPZs by granting such treatments also to domestic sectors. Several other zone types exist, each with particular features. Incentives in those EZs generally involve non-tax benefits, such as good infrastructure and cheap utilities, as well as reductions in customs duties, income taxes and other (local) taxes and fees. 22 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS A recent IDB study uses micro data to evaluate Export Free Zones in Costa Rica, El Salvador and Dominican Republic, including the specific role of tax holidays. It finds that (i) the Zones generally favor high profitability projects that would likely have gone ahead also without the incentives; (ii) projects are readjusted to keep incentives over time; and (iii) global tax avoidance is facilitated through the use of transfer prices vis-à-vis subsidiaries in the Zones (Artana, 2015). Life-cycle stages After approval, the tax administration should continue monitoring firms. This is an often neglected stage in many countries. Even when granted a tax holiday, taxpayers should still be required to file a tax return so that the authorities can assess the revenue cost of the incentive (as a central element of its tax expenditure review). Tax authorities should also periodically carry out audits to ensure that tax incentives are not abused. Conditions attached to incentives often require ongoing monitoring—for example, requirements that a given number of jobs be maintained, or that a certain percentage of production be exported (which may call for specialized tax forms for firms enjoying such incentives). Where non-compliance or abuse is detected, a penalty should be imposed and, ultimately, tax privileges denied. Making tax incentives temporary rather than permanent can have some attraction. One major attraction of a temporary tax incentive is that its expiry provides for a natural point of evaluation, feeding into a periodic reconsideration of whether the incentive should be continued, reformed or repealed. Temporariness of a tax incentive can also be used as a counter-cyclical policy. Indeed, when foreseen to be phased out in the near future, the investment effects of an incentive tend to be bigger than of permanent incentives (US Department of Treasury, 2010). Sunset provisions should be built into the law. In the absence of such a provision, firms may seek to roll-over a tax holiday, either by negotiating a new holiday period or by incorporating a new firm that may qualify for it. Tax holidays then become de facto permanent tax exemptions—a practice that should be avoided. B. Governance Good governance requires that the government’s decision-making process, its policies and the administration be transparent and subject to scrutiny and evaluation, to ensure that authorities can be held accountable for their actions and remedial action taken when necessary. This limits the scope for corruption, strengthens the trust of investors in government, and enhances confidence of the public that the tax system is fair in design and implementation. Transparency is a necessary but not a sufficient condition for good governance (see Box 3). IOs have developed a variety of practical tools to assess the transparency and wider governance performance of countries with respect to tax incentives.21 The background document presents templates that can be used as assessment tools. In general, the key requirements for good governance are that: 21 See for instance the IMF Fiscal Transparency Code (2014), including pillar IV on resource revenue management; IMF Manual on Fiscal Transparency (2007); OECD Best Practices for Budget Transparency (2002). The background document presents recent templates developed by the OECD and the World Bank. 23 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS  The award and monitoring of incentives be guided by the rule of law, with clarity about eligibility criteria;  Authority to grant national-level tax incentives (related to national taxes) rest solely with the Ministry of Finance; decisions drawing, as need be, on the views of stakeholders across government;  There be effective and transparent administration and evaluation. Box 3. Transparency Transparency is fundamental to empowering all stakeholders—the legislature, businesses, civil society and the public at large—with information about tax incentive policies, so that they can hold government accountable for its decisions. Transparency also generates information required for evaluation. Transparency should be created along three dimensions:  Legal: tax incentives should have a statutory basis in relevant tax laws.  Economic: the rationale for tax incentives should be clearly spelled out to enable a public debate on the country’s policy priorities. The costs and benefits of an incentive scheme should be assessed ex-ante and ex-post, based on clearly stated assumptions and methodologies, and the assessments published.  Administrative: qualifying criteria should be clear, simple, specific and objective, so as to reduce the discretion afforded to officials that grant the incentives. The decision-making process should be open and a list of incentives granted should be published. Because the revenue costs of tax incentives are usually not obvious, governments tend to face limited scrutiny when granting them, in contrast to the case of direct expenditures. Budgeting the amount of revenue forgone from tax incentives by the Ministry of Finance and revealing it to the public supports good governance, enables informed budget making and improves accountability. Transparency in the benefits they receive should also be required from multinational corporations. This can strengthen corporate social responsibility as employees, customers and local communities can hold corporations accountable for their decisions, including their tax treatment. 24 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Rule of law Tax incentives should be clearly prescribed in the law… This ensures that the legal basis governing the tax incentive has been approved by the legislature and has passed appropriate parliamentary and public scrutiny. If, on the other hand, tax incentives are negotiable and provided through decrees, agreements, regulations and the like, they escape oversight and can become prone to undue influence. Tax incentives in Nigeria can be introduced through laws, budget speeches, government notices/directives, executed agreements, as well as Memoranda of Understanding between the government and businesses (Nlerum, 2011). …and preferably consolidated in the tax law. The transparency and accessibility of tax incentives are compromised when embedded in multiple pieces of legislation applying, for instance, to different sectors. Tax incentives are for this reason best consolidated into the main body of the tax law. This reduces the likelihood of conflicting or overlapping provisions, which could create unintended distortions and uncertainty, as well as revenue losses. In practice, however, more than half of all LICs provide tax incentives outside the tax law (Figure 4). Figure 4. Governance of tax incentives by income group/1 80% 70% 60% 50% 40% 30% 20% 10% 0% Low Income (32) Low Middle (37) Upper Middle (43) High Income (42) Discretionary process available Incentives outside Tax Laws /1 Percentage of countries in four income groups that have discretion/provide incentives outside the tax law. Discretion is defined as processes under which the taxpayer can negotiate special incentives, or where approval of a tax incentive can be granted outside of the tax/customs agency. Source: Calculations based on James (2014) 25 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS In December 2012, special tax measures in as many as 17 laws and legislative acts were either abolished or consolidated into the General Tax Code in Senegal, significantly improving transparency of the tax system. The comprehensive tax reform, along with tighter administrative measures, streamlined the tax system and represented a significant rollback of tax incentives and exemptions. (Journal Officiel de la République du Sénégal, 2012) Eligibility criteria for granting tax incentives should be clearly defined and readily verifiable to allow for a rules-based approach. The law (and its accompanying regulations) should specify the conditions that the taxpayer needs to satisfy in order to qualify for a tax incentive, with as little room as possible for subjective interpretation or negotiation. Granting incentives can then be largely automatic by verification of stipulated criteria. Yet, not all tax incentives can be granted automatically, for instance because the law cannot always prescribe eligibility under all possible circumstances. This often introduces some elements of discretion. The extent of discretion, however, should be kept to a minimum as it introduces the risk of rent-seeking behavior on the part of the investors and corrupt behavior on the part of the public officials—a major concern in many countries.22 Excessive discretion can also serve as a signal of poor governance and thus alarm new investors. Figure 4 shows that, troublingly, close to 70 percent of LICs allow for granting processes that are largely discretionary in nature. There are several examples of considerable discretionary interpretation. For instance, the recently promulgated Foreign Investment Law of Myanmar loosely defines eligibility for tax exemptions over a “suitable period” for businesses that are “beneficial for the State.”(Myanmar Foreign Investment Law, 2012). The Gambia’s investment promotion authority confers a special status on investors, which are then awarded special investment certificates entitling them to benefit from incentive packages; the investment promotion authority states that, “Apart from these specific incentive packages, others can be negotiated with the Agency depending on the strategic nature of the investment.” (James, 2014). Tanzania’s ‘Strategic Investor Status’ allows companies investing over US$ 20 million to negotiate individual tax breaks. These “special concessions to individual companies … have never formally been made public” (Tax Justice Network, 2012). Haiti’s Investment Code covers virtually any economic activity and provides no selection criteria to be applied by the inter-ministerial committee in charge of granting discretionary incentives, which include both a 15-year CIT holiday and accelerated depreciation (Investment law of Haiti, 1989). There is reason for optimism too, as increased public awareness and civil society activism have urged governments to bring some previously secret deals into daylight. In 2012, the government of Sierra Leone established an online database for mining contracts, to make information on natural resource extraction contracts publicly accessible. Similarly, dozens of 22 The World Bank defines corruption as the abuse of public office for private gain; rent seeking reflects the use of companies’ resources to obtain economic benefit without wealth creation. Both can hamper development, see e.g. Abed and Gupta (2002). 26 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS previously secret extractive industry contracts were placed online by the government of Guinea (Africa Progress Panel, 2013). Authority to grant It is critical that ministries and agencies involved in the granting of tax incentives coordinate their activities. Several government agencies are often involved in the foreign investment process, such as the Investment Promotion Agency, the Ministry of Economy and Ministries responsible for specific sectors, such as Agriculture, Tourism or Mining. These different players often bring specific expertise, which can be useful in the design of tax incentives or required in the evaluation of eligibility criteria. But they usually also have different objectives. For instance, investment promotion agencies often support tax incentives in order to attract investors, having little direct concern for the revenue consequences. The Ministry of Finance, in contrast, will emphasize that revenue needs to be raised to provide public goods, including key pillars of a business-enabling environment such as infrastructure. Where various Ministries do not properly coordinate and responsibilities are not centralized, incentives may overlap, be inconsistent, or even work at cross-purposes. Effective coordination is a daunting but critically important task. The ultimate and sole authority to enact tax incentives at the national level should be with the Minister of Finance. Countries that have been successful in attracting investment have generally adopted a holistic approach that places tax policy in the context of a broader national development strategy. It is common practice to have an interdepartmental adjudication committee with combined expertise that makes recommendations to the Minister of Finance about tax incentives. The latter should take the final decision to enact tax incentives and be responsible for their implementation through, or working closely with, the tax administration. Indeed, the Minister of Finance is best able to weigh different priorities while also keeping an eye on the cost of incentives. Where authority is outside the Ministry of Finance, special interests can easily dominate the general public interest. As many as 10 organizations/agencies have the authority to grant tax incentives and exemptions in Ghana, including the Parliament, Ministry of Finance & Economic Planning, Revenue Agencies, Minerals Commission, Environmental Protection Agency, Food and Drugs Board, Ghana Free Zones Board, Ghana Investment Promotion Council, and Ghana National Petroleum Company (Amegashie, 2011). The Solomon Islands recently implemented a comprehensive statutory regime, prescribing the process of granting tax exemptions for import duties, excise, Goods and Services Tax (GST), income tax, sales tax and stamp duties. If a request is received for a tax exemption, a formal “Exemption Committee” provides advice to the Minister for Treasury and Finance. The Committee consists of representatives of the revenue administration, the Ministry of Development and Planning and the Ministry of Commerce. The law also prescribes clear and public criteria for the award of tax incentives and demands publication of both the application and the decision regarding the incentive. (Revenue and Customs Exemption Committee Solomon Islands, 2013). 27 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Administration The revenue administration should be in charge of the implementation and enforcement of tax incentive schemes, as it has the unique authority, expertise and experience necessary for the execution of the tax law of which incentives should be part. Where tax incentives are simple and unambiguous, they might be self-assessed by the taxpayer and subject to ordinary control and auditing procedures. Many incentive provisions, however, require some form of approval by the tax administration. When verifying facts, information or certification may be needed from other specialized government agencies or ministries. For the tax administration, documentation and publication of the decisions is a prerequisite to ensure transparency. This enables it to be held accountable by government and taxpayers. It also enables government (preferably the Ministry of Finance) and other organizations to evaluate the costs and benefits of tax incentives. C. Reform IOs have frequently advised countries to remove, or radically adjust, their tax incentives, as they have often been found to be ineffective or inefficient, or governance structures were ill- performing. Yet, reform of tax incentive regimes has proven difficult. This is because tax incentives are usually not only driven only or even mainly by well-articulated economic concerns—aimed at improving the wellbeing of citizens—but also by political motivations. For instance, politicians may find it attractive to introduce new tax incentives to reveal their proactive stance in addressing weak economic performance, or to favor particular regions. The fiscal costs of granting those incentives, moreover, are usually not transparent and only arise as forgone revenue in the future. Tax incentives also create vested interests among businesses and within government, making them difficult to repeal, even if they are ineffective. Indeed, economic elites are sometimes able to influence the governance and design of tax incentives and shape public opinion where it suits their objectives. When successful, they become powerful lobbyists who can capture the political process to resist change (Moe, 2005). Yet, successful reforms have been achieved: In 2013, Jamaica undertook a major tax reform, eliminating many of its generous and discretionary tax incentives. The Minister of Justice said: “The Jamaican economy has not been well served by the existing regime of sector based incentives. The consensus is that such incentives may have been partly responsible for Jamaica’s lackluster record of growth by encouraging the misallocation of limited economic resources in our country.” (James, 2013) In 2009, India’s Ministry of Finance released a draft Direct Taxes Code and a discussion paper on the bill, which recommended that India should move away from profit-based tax incentives in favor of expenditure-based tax incentives. A final bill was sent to Parliament in 2010, and passed to a Standing Committee on Finance (SCF), which released its report in March 2012. The SCF report concurred with the Ministry of Finance that profit-based incentives facilitate artificial tax planning, and that this form of incentive should no longer be granted (while existing tax holidays previously granted for firms in special economic zones would continue to be respected) (Standing Committee on Finance 2011-12). 28 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Egypt passed a new income tax law in mid-2005 that provided for the phasing out of tax holidays while grandfathering current beneficiaries. Between 2005 and 2006, FDI into Egypt doubled. (Keen and Mansour, 2010) In 2006, Mauritius normalized the taxation of its export processing zone companies with that in other sectors and removed all provisions relating to tax credits and tax holidays (except for a four-year income tax holiday for small business). At the same time, the corporate tax rate was gradually reduced from 25 to 15 percent in 2008. (Keen and Mansour, 2010) In his latest budget speech, the Finance Minister of India proposed to reduce the corporate tax rate from 30 to 25 percent over the next 4 years, accompanied by rationalization and removal of various tax incentives and exemptions, which, he argues, have led to pressure groups, litigation and loss of revenue. The Minister expects base broadening and rate reduction to lead to higher levels of investment and growth and more jobs (Government of India, Ministry of Finance, February 28, 2015). When reforming tax incentives, governments should seek a balance between tax stability for incumbents and equal treatment of entrants to the market. Sunset provisions are often provided when incentives are scaled back due to policy reform. For instance, in the case of repeal of a tax holiday, the expectation can be that existing privileges will continue for incumbent firms. Investment laws sometimes even contain stability clauses for investors against adverse legislative changes, as a sign of government’s commitment. Such stability provisions, however, create an uneven playing field between old and new investors and can lead to significant distortions. Such situations should not last for too long. Government might therefore need to renegotiate existing incentive provisions or provide reasonable, time-bound incentives to new investors. INTERNATIONAL COORDINATION Tax incentives are often instances of tax competition, with the risk that ultimately all countries will lose from their use.23 By attracting investment that would otherwise have gone elsewhere, tax incentives can have adverse cross-border spillover effects on the welfare of other countries. To the extent that the allocation of investment is driven by tax rather than commercial considerations, the result is an inefficient global allocation of capital, and a consequent collective loss of output. Moreover, tax incentives in one country might evoke strategic reactions that lead them to offer similar policies (IMF, 2014). This process of tax competition can cause a race to the bottom, with all countries ultimately ending up with lower tax revenue and with no discernible impact on the allocation of investment. , 23 Tax competition might not be bad per se. For instance, some see tax competition as welcome in counteracting a political bias towards excessive public expenditure. Such arguments are less pervasive in LICs, however, which generally face an urgent need for higher domestic revenues to finance projects for development. 29 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Examining data for 40 countries in Latin America, the Caribbean and Africa, it appears that the length of a country’s tax holidays responds positively to that of its neighbors (Klemm and Van Parys, 2012). A race to the bottom is evident among special regimes in Africa, where effective tax rates have fallen to almost zero in industries where special regimes are in place (Abbas and Klemm, 2013). Grenada announced in 2012 that, in order to encourage the resort group Sandals to invest in his country, the government agreed to waive the company’s payment of corporate taxes for 25 years, place a cap on Sandals’ property taxes for 25 years and waive all import duties for 25 years. The minister announced that the benefits to the economy were an injection of US$100 million and the creation of 425 jobs. This reflects the aggressive nature of tax competition in the hotel industry in the Caribbean (James, 2013). Tax coordination offers opportunities to address spillovers—but also involves the risk of inducing other harmful responses. In principle, a coordinated response can prevent mutually harmful outcomes induced by uncoordinated tax design, for instance by agreeing to a ban on the use of certain tax incentives. Most forms of international coordination, however, are only of limited scope and scale. This raises two fundamental issues that require careful consideration when assessing the impact of tax coordination:  Coordination with respect to some tax incentives can intensify tax competition in other provisions. Suppose for instance that countries agreed not to offer special treatment to mobile activities (which, as discussed above, they might wish to do) but could otherwise compete on the general rate of tax they apply. The outcome could then be worse for all than if differential treatment were allowed, since they will then compete more aggressively on the rules covering all industries, including those with immobile capital (see e.g. Keen, 2002). Against this, of course, must be weighed the disadvantages of differential treatment across sectors discussed above.  Coordination among a subset of countries can intensify tax competition with outsiders— with the latter the major beneficiaries. Coordination is best achieved among the countries that most directly compete for mobile capital. If too limited in regional scope, the tax base of the participating countries can become more vulnerable to pressures from outside jurisdictions with lower taxes: they are in effect setting themselves up for increased competition from non- participants, who may consequently be the major beneficiaries. The scope of coordination agreements should thus be sufficiently broad in terms of country coverage,24 the difficulty being that of overcoming the incentive to benefit by staying outside the agreement. International tax coordination can take different forms. For instance, countries can agree on a non-binding code of conduct not to use certain tax incentives, such as tax holidays—as for instance has been done with the Code of Conduct for business taxation in the European Union. Something 24 Or other relevant characteristics. Similar arguments apply, for instance, to cooperation between producers of some natural resource, whatever their geographical location. 30 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS more binding would be a common legislative framework, akin to the state-aid rules in the European Union. Yet, tax coordination has proven difficult in practice in many regional groupings. Negotiating and implementing an agreement is a substantial effort, and will require an effective supranational monitoring framework and powerful institutions to enforce it—something that is lacking in many regions. More realistic in these circumstances might be to start with a more modest form of cooperation. For instance, countries could agree on a common framework for reporting tax incentives and information exchange to encourage mutual learning. This could also enhance transparency and governance practices, and enable ex-post assessment of tax incentives. The Southern African Development Community (SADC) aims to reduce and ultimately to eliminate tax competition that damages the region’s revenue mobilization efforts. The SADC Protocol on Finance and Investment provides for co-operation and coordination on tax incentives in the region. The “Guidelines for the application and treatment of tax incentives in the SADC region” seeks endeavors by the Member States to avoid harmful tax competition or introducing tax legislation that prejudices another Member State’s economic policies or activities (SADC, 2012). The members of the East African Community (EAC) have recently made progress towards coordination of their tax incentive regime through the use of a ‘Code of Conduct’, though this is yet to be adopted. This aims to formalize an existing arrangement under which each year the EAC finance ministers meet before their budget speeches are made and discuss their budget proposals. This provides the opportunity for Finance Ministers to dissuade other members if they propose any new tax incentive that puts other countries at a disadvantage (Tax Justice Network, 2012). In the West African Economic and Monetary Union (WEAMU), considerable efforts have been made to set-up a structure to tackle the problem of tax competition by issuing directives that limit the applicable tax rates that countries can use. The coordination framework has led to some convergence of countries’ tax systems, and in turn to positive revenue effects in WAEMU member states. However, there are large gaps between de jure and de facto coordination, as WAEMU has failed to provide its regional institutions with the necessary resources to undertake effective surveillance, which has led to ineffective enforcement and undermined the credibility of coordination. In fact, the framework allows for unfettered tax competition as long as this is done outside the countries’ main tax laws. This has made their tax systems opaque, increased complexity and contributed to a culture of tax negotiation (Mansour and Rota-Graziosi, 2013). In Central America, coordination of tax incentives has some history. Between 1962 and 1984, four countries (Costa Rica, El Salvador, Guatemala and Nicaragua) were parties to a convention on tax incentives, which mandated uniform tax incentives and prohibited the introduction of new ones. In 2012, the Committee of Ministers of Finance of Central America, Panama, and the Dominican Republic (COSEFIN) discussed the adoption of a “Declaration of Good Practices” for investment tax incentives, which was preceded by a wide-ranging exercise to quantify the cost of existing tax incentives to make a stronger case for their removal. Discussions on a mandatory code of good practices remain ongoing (Cebotari et al., 2013). 31 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS International tax rules adopted by advanced and emerging economies may affect the intensity of tax competition among LICs. The benefits of tax incentives to investors are larger, for instance, if parent companies reside in countries with territorial taxation and/or weak CFC regimes, since (as discussed above), because the exemption of qualified foreign-earned income in the residence country preserves the benefit of tax advantages in the host country. Advanced and emerging countries with worldwide taxation, however, sometimes offer tax sparing clauses, as discussed in footnote 14. This too increases the potential effectiveness of tax incentives. This in itself, however, is likely to intensify tax competition among LICs to attract FDI. Ultimately, territoriality and tax sparing may therefore not make LICs collectively better off. Tax incentives as part of bilateral or multi-party investment agreements vary widely in form and there is no standard template for them. There are often exceptions providing that tax matters are not covered by such agreements. This might refer to tax matters generally, those confined to double tax treaties (not extending to indirect taxes such as VAT), or those related to specific investment treaty obligations, such as the "most favored nation" provision. Where a tax incentive is provided, it is important to consider whether the relevant investment treaties might unintentionally be automatically accorded to other taxpayers, not originally intended to benefit from the tax incentive (such as under a most favored nation clause). Moreover, protections under the investment treaty (such as stabilization clauses) may make it hard to withdraw the tax incentive when it no longer serves the original purpose, or may require compensation not otherwise payable. Also important is which dispute settlement procedure would apply if tax issues fall under the investment treaty (usually there is a binding arbitration clause), and what should be done to prepare for such a possibility. These issues are best considered in collaboration with those responsible for negotiating the investment treaties. CONCLUSIONS This paper finds that many LICs have considerable scope to improve the effectiveness and efficiency of their investment tax incentives. The paper has set out options for improvement in design and governance, at both national and international levels. At the national level, there is generally scope to improve the design of tax incentives (for example by placing greater emphasis on cost-based incentives rather than profit-based ones; and by targeting tax incentives better), strengthen their governance (for instance through more transparency, better tax laws and a stronger role of the Minister of Finance) and by undertaking more systematic evaluations. At the international level, countries may gain by coordinating their tax incentive policies regionally, so as to mitigate the negative spillovers from tax competition. Making progress requires concerted action from many stakeholders. National reform and international coordination of investment tax incentives have often proven difficult to achieve for a variety of reasons. Making better progress requires that political decisions about tax incentives be based on proper analysis of their effectiveness and efficiency, which requires transparency along with systematic information gathering and evaluation. Stakeholders need to accept their various responsibilities if progress is to be made: 32 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS  Governments in LICs bear the prime responsibility for the design and governance of tax incentives. It is critical that they can be held accountable by parliament, businesses, donors, relevant regional and international bodies, and the public at large. This requires transparency, as well as a properly functioning legal system. Also important is that governments be equipped with the necessary skills, tools and data to undertake evaluations, to enable informed decision making. Publishing an annual tax expenditure review as part of the budgetary process is an important step in this direction.  Society at large, including civil society organizations (CSOs) and media, play a vital role in scrutinizing government decisions, detecting malfunctioning practices, and mobilizing voice. They can, for instance, draw attention to and publicize the revenue foregone from tax incentives that is reducing the resources available for development.  Corporations, including foreign-owned multinationals that enjoy tax incentives bear corporate social responsibility, including towards (local) communities, governments, employees, consumers and other stakeholders. This means, for instance, that they should comply not only with the letter but also the spirit of the tax law. Going further, the business sector might develop and subscribe to a code of conduct under which incentive packages that fail to meet certain minimum transparency standards are not accepted or sought.  Supranational regional bodies have delegated responsibility from national governments to enforce commitments agreed upon in the cooperation. Their powers will be as strong as the members want them to be. Supranational bodies can support transparency by developing common standards for reporting, data collection and evaluation of tax incentives.  Governments in donor countries, including those in the G20, have a responsibility in: (i) the design of their own corporate tax policies, to avoid harmful spillovers onto LICs (for example, through treaty abuse; (ii) the provision of donor aid, which itself in many cases enjoys the benefits of tax incentives offered by LICs (a practice for which the rationale is increasingly unclear, and which is being revisited by several donors);25 (iii) imposing conditionality regarding tax incentives when providing donor aid, such as meeting minimum transparency standards.  International organizations, such as those co-authoring this document, need to build on their established experience to support their LIC members in building more efficient and effective tax system. This includes support in developing their capacity to employ tools to better analyze tax incentives (such as micro simulation models and effective tax rate models), and evaluations that help inform the debate (such as tax expenditure reviews, governance assessment and more comprehensive cost-benefit assessments)—as discussed in detail in the background document. 25 See e.g. International Tax Dialogue (2006). 33 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Realizing improvements in effectiveness and efficiency in the use of tax incentives is important for LICs to achieve better development outcomes. Higher investment can then be reconciled with more abundant domestic revenue mobilization, much-needed for public spending programs and established as a development priority in the Addis Ababa Action Agenda.26 26 Draft Resolution by the General Assembly endorsing the Addis Abeba Action Agenda A/69/L.82. 34 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Appendix I. Consultations This paper draws on the extensive technical assistance provided by the IMF and World Bank on tax incentive issues in developing LICs over many years, and their continuing dialogue with them on tax matters. In preparing this paper, the IOs also held several discussions during public consultation meetings with representatives from developing countries, civil society organizations, businesses and academics. Events were organized during March and April 2015 in Paris, Washington DC and New York. In response to an online invitation for comments on a draft version of the paper during July and August 2015, several submissions were received. A short review of key issues raised is listed below. Developing country officials showed great interest in deepening their analysis of tax incentives with the tools described in the paper, including for the analysis of tax holidays and economic zones. Most officials noted that their countries have not carried out any in-depth analysis of the costs and benefits of tax incentives, yet are frequently confronted with pressures from businesses to grant them. For the fear of losing investments, generous tax incentives are then offered with ample tax receipts foregone. In some regions in Africa, countries are coordinating efforts to assess the revenue costs of tax incentives—sometimes showing very large numbers. However, methodologies are not harmonized, which hampers comparison between studies. Civil society expressed the view that tax incentives are often ineffective and inefficient. They also emphasized the importance of ‘spillover effects’ of tax policies in one country on other countries and therefore the need for international cooperation. They argue that developing countries are particularly vulnerable to tax incentives, reducing the domestic resources for their own development. Organizations also noted the inherent difficulties of performing a proper cost-benefit assessment, yet urged the IOs to support such exercises in LICs so as to improve information in the decision- making process. Finally, the importance of transparency—from both governments and businesses— was emphasized. It was argued that G20 countries need to work together with LICs to develop a global coordinated approach, with developing counties having an equal say at the negotiation table. Business argued that tax incentives can be effective instruments to stimulate investment, but only when used as part of a well-conceived and well-implemented strategy. Thus, tax incentives require clear pre-established objectives, should be based in the law and efficiently managed. They emphasized that it is essential for investment that the tax regime be predictable, efficient and stable. Business also argued that tax incentives cannot compensate for an otherwise poor tax structure or for an unfavorable investment climate, such as significant administrative barriers, limitations to competition, no independent legal system, or failing property rights protection. Finally, they find that countries should be able to set their own tax policies and allocate their own tax incentives, since there is no one-size-fits all. Regional coordination is not considered to be helpful. Independent contributors offered important information and ideas. One contribution offered insight into tax incentives in ten ASEAN countries. It shows that, while 10 countries grant tax holidays, 6 of them also allow for their renewal upon expiration. Contributors also warned against 35 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS the outcomes from investor surveys, which may give a biased view regarding the actual behavior of investors. An important point raised was the lack of relevant data and the need for more systematic data gathering by tax administrations. This will require them to monitor firms, even if they have been granted tax exemptions. Also emphasized was the need to explore the effectiveness of tax incentives under the unique circumstances of every country. Others noted that tax incentives are not only prevalent in LICs, but are widespread in advanced countries. Several contributors emphasized the importance of tax expenditure budgeting for enhancing transparency. 36 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS References Abbas, S. M. Ali, and Alexander Klemm, 2013, “A partial race to the bottom: corporate tax developments in emerging and developing economies”, International Tax and Public Finance, Vol. 20, pp. 596–617. Abed, George T. and Sanjeev Gupta, 2002, “Governance, Corruption and Economic Performance”, International Monetary Fund, Washington D.C. Adams, Samuel, 2009, “Foreign Direct Investment, Domestic Investment, and Economic Growth in Sub-Saharan Africa”, Journal of Policy Modeling, Vol. 31, pp. 939–949. Africa Progress Panel, 2013, “Africa Progress Report Equity in Extractives: Stewarding Africa's Natural Resources for All.” Aitken, Brian J., and Ann E. Harrison, 1999, “Do Domestic Firms Benefit from Foreign Direct Investment? Evidence from Venezuela”, American Economic Review, Vol. 89, 605-618. Alder, Simon, Lin Shao and Fabrizio Zilibotti, 2013, “The Effect of Economic Reform and Industrial Policy in a Panel of Chinese Cities”, Center for Institutions, Policy and Culture in the Development Process, Working Paper 207, University of Zurich. Amegashie, Atsu, 2011, “Ghana’s Regime of Exemptions from Taxes and Duties: Guidelines for Reform,” Department of Economics, University of Guelph, Canada. Artana, Daniel, 2015, “The effectiveness of fiscal incentives: The case of the export free zone of Costa Rica, El Salvador and Dominican Republic, in Osmel Manzano, Mario Cuevas and Sebastián Auguste (eds.), Partners or Creditors? Attracting foreign investment and productive development to Central America and Dominican Republic,” Inter American Development Bank. Auriol, Emmanuelle, and Michael Walters, 2012, “The Marginal Cost of Public Funds and Tax Reform in Africa”, Journal of Development Economics, Vol. 97, pp. 58-72. Bird, Richard M., 2000, “Tax Incentives for Investment in Developing Countries,” in Fiscal Reform and Structural Change in Developing Countries, ed. by Guillermo Perry, John Whalley, and Gary McMahon, Vol. 1, pp. 201-221. Bird, Richard M. and Eric M. Zolt, 2008, "Tax policy in emerging countries", Environment and Planning C: Government and Policy, Vol. 26, pp. 73-86. Blomstrom, Magnus and Ari Kokko, 1998, “Multinational Corporations and Spillovers”, Journal of Economic Surveys, Vol. 12, pp. 247-277. 37 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Boadway, Robin W. and Anwar Shah, 1992, Perspectives on the Role of Investment Incentives in Developing Countries (Washington: World Bank). Bolnick, Bruce, 2004, “Effectiveness and Economic Impact of Tax Incentives in the SADC Region”, USAID, Technical Report. Bwalya, Samuel Mulenga, 2006, "Foreign direct investment and technology spillovers: Evidence from panel data analysis of manufacturing firms in Zambia," Journal of Development Economics, Vol. 81, pp. 514-526. Cebotari, Aliona, Melesse Tashu, Selcuk Caner, Denise Edwards-Dowe, Brian Jones, Vinette Keene, Robert Mills, and Sumiko Ogawa, 2013, “Enhancing Fiscal Revenue,” in The Eastern Caribbean Economic and Currency Union: Macroeconomics and Financial Systems, ed. by Alfred Schipke, Aliona Cebotari, and Nita Thacker (Washington: International Monetary Fund). Cheng, Leonard K., and Yum K. Kwan, 2000, “What are the Determinants of the Location of Foreign Direct Investment? The Chinese experience,” Journal of International Economics, Vol. 5, pp. 379– 400 Clark, W. Steven, 2000, “Tax Incentives for Foreign Direct Investment: Empirical Evidence on Effects and Alternative Policy Options,” Canadian Tax Journal, Vol. 48, pp. 1139-1180. De Mooij, Ruud A., and Sjef Ederveen, 2008, “Corporate Tax Elasticities: A Reader’s Guide to Empirical Findings,” Oxford Review of Economic Policy, Vol. 24, pp. 680-97. Du, Luosha, Ann Harrison, and Gary Jefferson, 2011, “FDI Spillovers and Industrial Policy: The Role of Tariffs and Tax Holidays,” NBER Working Paper No. 16767 (Cambridge, Massachusetts). Dunning, John H., 1993, Multinational Enterprises and the Global Economy, Harlow, Addison Wesley. Easson, Alexander J., 2001, “Tax Incentives for Foreign Investment, Part I, Recent Trends and Countertrends,” Bulletin for International Fiscal Documentation, Vol. 55, pp. 266-274. Easson, Alexander J., 2004, Tax Incentives for Foreign Direct Investment, Kluwer Law International. Easson, Alexander J. and Eric M. Zolt, 2004, “Tax Incentives”, (Washington: World Bank Institute). Ebrill, Liam, Michael Keen, Jean-Paul Bodin, and Victoria Summers, 2001, “The Modern VAT”, International Monetary Fund. Economist, The, 2015, Special Economic Zones: Not so Special, April 4, 2015. 38 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Girma, Sourafel, Holger Görg and Mauro Pisu, 2007, “Exporting, Linkages and Productivity Spillovers from Foreign Direct Investment,” Canadian Journal of Economics, Vol. 41, pp. 320-340. Gorodnichenko, Yuriy, Jan Svejnar, and Katherine Terrell, 2007, “When Does FDI Have Positive Spillovers: Evidence from 17 Emerging Market Economies,” IZA Discussion Paper No. 3079. Government of India, Ministry of Finance (2015), Budget 2015-2016 Speech, New Delhi, February 28. Government of Senegal, General Secretariat (2012), Official Journal, Dakar, January 3. Griffith, Rachel, Stephen Redding and John van Reenen, 2004, “Mapping the Two Faces of R&D Productivity Growth in a Panel of OECD Countries,” The Review of Economics and Statistics, Vol. 86, pp. 883-895. Grubert, Harry, and John Mutti, 2004, “Empirical Asymmetries in Foreign Direct Investment and Taxation,” Journal of International Economics, Vol. 62, pp. 337-58. Gugl, Elizabeth and George Zodrow, 2006, “International Tax Competition and Tax Incentives in Developing Countries,” in The Challenge of Tax Reform in a Global Economy, ed. by James Alm, Jorge Martinez-Vazquez, and Mark Rider (Berlin: Springer). Haddad, Mona and Ann Harrison, 1993, “Are their Positive Spillovers from Foreign Direct Investment?” Journal of Development Economics, Vol. 42, pp. 51-74. Haskel, Jonathan E., Sonia C. Pereira and Matthew J. Slaughter, 2007, “Does Inward Foreign Direct Investment Boost the Productivity of Domestic Firms?”, The Review of Economics and Statistics, Vol. 89, pp. 482-496. IMF, 2007, “Manual on Fiscal Transparency”, Washington. IMF, 2012, “Fiscal Policy and Employment in Advanced and Emerging Economies”, Policy Paper (June 15, 2012). IMF, 2014, “Spillovers in International Corporate Taxation”, Policy Paper (May 19, 2014). IMF, OECD, UN and World Bank, 2011, Supporting the Development of More Effective Tax Systems, Report to the G-20 Development Working Group. Indian Minister of Finance (Arun Jaitley), 2015, Budget 2015-2016, Presented on February 28, 2015, see http://indiabudget.nic.in/bspeecha.asp. International Tax Dialogue, 2006, Tax treatment of donor-financed projects, October 3 2006. 39 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS James, Sebastian, 2007, “The Effect of Tax Rates on Declared Income: An Analysis of Indian Taxpayer Response to Changes in Income Tax Rates,” Ph.D. diss., Harvard University, Cambridge, Mass. James, Sebastian, 2013, “Investment Incentives in Other Geographies”, mimeo, World Bank. James, Sebastian, 2014, “Tax and non-tax incentives and investments: Evidence and Policy Implications”, Investment Climate Advisory Services. World Bank Group, June 2014. James, Sebastian, and Stefan Van Parys, 2009, “Investment Climate and the Effectiveness of Tax Incentives,” World Bank Group. Javorcik, Beata, 2004, “The Composition of FDI and Protection of Intellectual Property Rights: Evidence from Transition Economies,” European Economic Review, Vol. 48, pp. 39-62. Keen, Michael, 2002, “Preferential Regimes Can Make Tax Competition Less Harmful,” National Tax Journal, Vol. 54, pp. 757–762. Keen, Michael and Mario Mansour, 2010, “Revenue Mobilization in Sub-Saharan Africa: Challenges from Globalization II – Corporate Taxation,” Development Policy Review, Vol. 28, pp. 573-596. Keller, Wolfgang and Stephen R. Yeaple, 2003, “Multinational Enterprises, International Trade and Productivity Growth: Firm Level Evidence from the U.S.”, NBER Working Paper No. 9504 (Cambridge, Massachusetts). Kenneth, Thomas P., 2011, Investment Incentives and the Global Competition for Capital (New York: Palgrave Macmillan). Kinda, Tidiane, 2014, “The Quest for Non-Resource-Based FDI: Do Taxes Matter?”, IMF Working Paper no. 14/15 (Washington: International Monetary Fund). Klemm, Alexander, 2010, "Causes, benefits, and risks of business tax incentives." International Tax and Public Finance, Vol. 17, pp. 315-336. Klemm, Alexander, and Stefan Parys, 2012, "Empirical Evidence on the Effects of Tax Incentives," International Tax and Public Finance, Vol. 19, pp. 393-423. Lent, George E., 1967, “Tax incentives for investment in developing countries”, IMF Staff Papers Vol. XIV, pp. 249-323. Leong, Chee, 2013, "Special Economic Zones and Growth in China and India: An Empirical Investigation," International Economics and Economic Policy, Vol. 10, pp.549-567. 40 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Mansour, Mario and Grégoire Rota-Graziosi, 2013, "Tax Coordination, Tax Competition, and Revenue Mobilization in the West African Economic and Monetary Union," IMF Working Paper no. 13/163 (Washington: International Monetary Fund). McLure, Jr., Charles E., 1999, “Tax Holidays and Investment Incentives: a Comparative Analysis, Bulletin for Int’l Fiscal Documentation, Vol. 53, pp. 326-339. Mintz, Jack, 1990, “Corporate Tax Holidays and Investment,” World Bank Economic Review, Vol. 4, pp. 81-102. Moe, Terry M., 2005, “Power and Political Institutions”, Perspectives on Politics, Vol. 3, pp. 215-233. Myanmar Foreign Investment Law (FIL), The Pyidaungsu Hluttaw Law No. 21/2012. November 2012, Chapter XII, Exemptions and Benefits. http://export.gov/thailand/static/Foreign%20Investment%20Law_Latest_eg_th_055982.pdf Nlerum, Francisca E., 2011, “Reflection on the Attitude of the Courts to Tax Incentive Mechanism in Nigeria,” NIALS Journal of Business Law. Available on the internet: http://www.nials- nigeria.org/journals/Dr.Francisca%20E.%20Nlerumbus.pdf Organisation for Economic Co-operation and Development, 2001, “Corporate Tax Incentives for Foreign Direct Investment,” OECD Tax Policy Study No. 4. ———, 2002, Foreign Direct Investment for Development – Maximizing Benefits, Minimizing Costs. ———, 2006, “Tax Incentives for Investment – A Global Perspective: experiences in MENA and non- MENA countries” in Making Reforms Succeed: Moving Forward with the MENA Investment Policy Agenda. Papua New Guinea Taxation Review (2013-2015), Issues Paper No.5: An examination of the advantages and disadvantages of tax incentives, December 2014. Reside, Renato, 2006, “Fiscal Incentives and Investments in the Philippines”, Economic Policy Reform & Advocacy & United States Agency for International Development . Rodrik, Dani, 1993, “Taking Trade Seriously: Export Subsidization as a Case Study in Policy Effectiveness,” National Bureau of Economic Research, Working Paper No. 4567. SADC– Trade, Industry, Finance and Investment (TIFI) Directorate, 2012, Striving for Regional Integration. Baseline Study on the Implementation of the SADC Protocol on Finance and Investment (http://www.giz.de/en/downloads/giz2012-en-implementation-sadc-finance- investment.pdf ) 41 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Schiffbauer, Marc, Abdoulaya Sy, Sahar Hussain, Hania Sahnoun and Philip Keefer, 2015, Jobs or Privileges: Unleashing the Employment Potential of the Middle East and North Africa, World Bank Mena Development Reports. Shah, Anwar, 1995, Fiscal Incentives for Investment and Innovation, Oxford University Press. Solomon Islands, 2013, Revenue and Customs Exemption Committee – Non-Statutory exemptions regulations 2013. Tanzi, Vito and Howell H. Zee, 2000, “Tax Policy for Emerging Markets: Developing Countries, Working Paper No. 00/35 (Washington: International Monetary Fund). Tanzi, Vito and Parthasarathi Shome, 1992, The Role of Taxation in the Development of East Asian Economies, in The Political Economy of Tax Reform, NBER-EASE Volume 1 Volume Author/Editor: Takatoshi Ito and Anne O. Krueger, editors Volume Publisher: University of Chicago Press. Tax Justice Network Africa & ActionAid, 2012, Tax Competition in East Africa: A Race to the Bottom? UNCTAD, 1996, Incentives and Foreign Direct Investment. UNCTAD, 2000, Tax Incentives and Foreign Direct Investment: A Global Survey. UNIDO, 2011, Africa Investor Report: Towards evidence-based investment promotion strategies. U.S. Department of the Treasury, 2010, “The Case for Temporary 100 Percent Expensing: Encouraging Business to Expand Now by Lowering the Cost of Investment – A Report by the U.S. Department of the Treasury’s Office of Tax Policy. Van Parys, Stefan, and Sebastian James, 2010, “The Effectiveness of Tax Incentives in Attracting Investment: Panel Data Evidence from the CFA Franc Zone,” International Tax and Public Finance 17, pp. 400-429. Wang, Jin, 2013, The Economic Impact of Special Economic Zones: Evidence from Chinese Municipalities, Journal of Development Economics 101, 133-147. Wooster, Rossitza B. and Diebel, David S., 2006, Productivity Spillovers from Foreign Direct Investment in Developing Countries: A Meta-Regression Analysis. Xu, B., 2000, “Multinational Enterprises, Technology Diffusion and Host Country Productivity Growth.” Journal of Development Economics, Vol. 62, pp. 477-493. 42 TAX INCENTIVES FOR INVESTMENT—OPTIONS FOR LICS Yudaeva, K. K. Kozlov, N. Melentieva and N. Ponomareva, 2003, “Does Foreign Ownership Matter? The Russian Experience.” The Economics of Transition, 2, pp. 383-410. Zee, H., J. Stotsky, and E. Ley, 2002, Tax Incentives for Business Investment: A Primer for Policy Makers in Developing Countries,” World Development, Vol. 30, pp. 1497-1516. Zeng, Douglas Zhihua, 2015, Global Experiences with Special Economic Zones, World Bank Policy Research Working Paper 7240. Zolt, Eric, 2014, Tax Incentives: Protecting the Tax Base, Paper Prepared for the UN Project: Protecting the Tax Base for Developing Countries. 43