Chapter

13. Tax competition and coordination in extractive industries

Editor(s):
Michael Keen, and Victor Thuronyi
Published Date:
September 2016
Share
  • ShareShare
Show Summary Details

1 Introduction

Tax competition has been studied extensively in the economic literature, mainly out of concern that countries may collectively1 lose revenues when they set their tax policies independently of each other – relative to setting them cooperatively. Little attention has been given, however, to the issue of tax competition and coordination specifically in the extractive industries (EIs), where appropriately designed taxation could be particularly efficient given a number of sector-specific characteristics.

First, locational rent associated with EIs is potentially very high. This is the excess return that investors may earn from bringing the resource to market, relative to the minimum return they require. It is location specific when investors have a chance to earn this rent (or a portion of it) only if they invest in the jurisdiction where the resource is located. The implication is that government can tax such rent with very little impact on investors’ behavior – and since investors tend to be foreigners, especially in developing countries, their welfare should be of little concern to governments. It might seem that given such rent, the sector would not be subject (at least not to the same extent) to the forces that tax competition exerts on other (more mobile) factors and activities. But if, for some reason, downward tax competition in EI does happen,2 the revenue loss, especially in developing countries, could be substantial. IMF (2012) reports that revenue from EIs can be very high, exceeding 10 percent of GDP in 22 countries; in Sub-Saharan Africa, increases in government revenue since the early 1990s have come largely from upstream EIs (Figure 13.1). The exhaustibility of natural resources means that governments have only one chance to get their tax policy right.

Figure 13.1Composition of tax revenue in Sub-Saharan Africa: resource vs. non-resource; 1980–2010

Second, EIs exhibit substantial up-front investments, especially in exploration – but also in development and capital outlays for extraction. Given the sunken nature of such investments, governments may be tempted to bid their taxes down to attract investment but later increase them to extract a higher share of revenue. The question arises as to how this “time-inconsistency” problem interacts with tax competition concerns and whether tax coordination can alleviate (or even worsen) it. The problem is compounded by the high uncertainty regarding investment in EIs (i.e. the quantity and quality of the resource that can be developed and price uncertainty on inputs and output) and by the asymmetric information between investors and governments, with the latter generally suffering an informational disadvantage.

This chapter discusses issues of tax competition and coordination in upstream EIs. Is tax competition in EIs happening? And which taxes3 does it affect? Should governments coordinate taxes affecting EIs? What form should this coordination take? These are difficult questions that the literature has not fully resolved. Our aim in discussing them here is modest; we seek to provide some guidance on whether these should be significant concerns for policy makers and, if so, possible ways to address them. We do not discuss certain forms of limited coordination in tax policy and administration, such as tax treaties (covered in Chapter 5, Daniel and Thuronyi, this volume), and, specifically for the EIs, common development zones (covered in Chapter 10, Cameron, this volume) and Chapter 11, Daniel, Veung and Watson, this volume).

The term “tax competition” encompasses many forms of competition among tax jurisdictions and within them: competition for cross-border shoppers (e.g. through consumption taxes); competition for mobile factors (e.g. real and financial capital through corporate taxes or labor through personal income taxes and payroll taxes); competition for voters (e.g. through lower or higher property taxes in states of a federation to attract people with a set of particular preferences for local public goods); and “vertical tax competition” between levels of governments in a federation for the same tax base (e.g. royalties and CIT could be imposed by both the central and state governments in a federation).

Tax competition could also have the objective of attracting a tax base, such as profits, without any impact on real activities.4 To limit our discussion, we define tax competition as “non-cooperative tax setting among jurisdictions with the aim to attract mobile factors of production (capital and labor) or firms in EIs”. We therefore exclude tax competition to attract profit shifting. Although these two categories of competition are difficult to disentangle, this exclusion should not be problematic for our purpose since resource-rich countries rarely set their tax rates so low that they are likely to attract paper profits. Resource-rich countries, especially developing, would probably lose from setting tax rates too low to compete for paper profits rather than setting them reasonably high to tax the resource rent.5

Under this definition, the major taxes that can be used by countries to compete include (1) CIT and other taxes on production, such as royalties and resource rent taxes (RRTs); (2) tariffs on imports of capital and intermediate inputs and tariffs on exports, which, although rare, are still important in some developing countries for EIs; (3) cascading sales taxes; (4) withholding taxes on payments to non-residents; and (5) labor taxes in general and additional taxes on expatriates in particular – which may take the form of fees for visas and work permits. These taxes have varying impacts on effective tax rates on capital and labor and hence may affect both the decision to locate and marginal decisions (e.g. to expand, or not, and exploration and production activities). To simplify the discussion, and also due to data constraints, we limit the analysis to the instruments in 1, 2, and 3.

The chapter is organized as follows. Section 2 tries to draw some lessons from the general literature on tax competition and coordination for the EI sector. Section 3 brings these to the specific context of the EI sector and attempts to answer some basic questions such as whether tax competition in EI is happening and what countries can or should do about it. Section 4 reviews experiences in tax coordination, drawing mainly from those in the European Union (EU) and the West African Economic and Monetary Union (WAEMU), and IMF technical assistance to developing and resource-rich countries, identifying along the way remaining challenges. Section 5 concludes.

2 Lessons from the literature on tax competition and coordination

The main issue that the tax competition literature addresses is the possibility that countries may be collectively worse off setting their tax policies alone rather than cooperatively because policy decision in one country have externalities (or spillovers) that affect the tax base of other countries. Hence, in the absence of tax coordination, tax rates may be set lower or higher than the optimal level dictated solely by national considerations. The prominent role of foreign direct investment in the resource sector, especially in developing countries, may dampen tax competition for lower rates since taxing foreigners should be less of a concern to governments than taxing their own citizens or companies. This may explain, at least in part, why we do not observe downward pressure on average effective tax rates in the resource sector (more on this later).

The literature has mushroomed in the past two decades as globalization (including the marked decline in tariff and non-tariff barriers to trade in goods and services, capital, and to a much lesser extent labor) exerted downward pressure on taxes on mobile factors and firms, which in turn raised concerns as to the viability of such taxes as sources of government revenue. The CIT has occupied center stage of the debate, but other taxes have attracted attention as well, such as those on labor, portfolio income, and consumption. At the same time, developing countries suffered significant loss of tariff revenues that some were not able to fully recover using domestic taxes.6 This section provides a selective review of major lessons from this literature, both theoretical and empirical, with an eye to the questions posed earlier. We also seek to clarify some of the misconceptions about tax competition, which often come up in IMF technical assistance work in tax policy.

Keen and Konrad (2012) provide one of the most recent and comprehensive accounts of the theoretical literature on tax competition (mainly capital tax competition). One main lesson from their work is that the real world is still far more complex than current economic models can handle, especially in terms of differences across countries – in, for instance, size, economic structures, institutions, the complexity of tax systems, and the arbitrage opportunities they create. They also conclude that the literature has not yet reached any definitive conclusion on whether tax competition is “bad” or “good”7 and not even how to identify the former type – since this is the part that policy makers would want to suppress through coordination. They do suggest, however, that enough progress has been made to provide some useful guidance on how countries may effectively coordinate their tax policies. For example, agreeing on minimum tax rates somewhat above the lowest set in a non-cooperative outcome is more likely to be effective than full harmonization of taxes – in simple models at least, it produces a better revenue outcome for all participating countries (including that obliged to raise its tax rate) and is less stringent than full harmonization.8 But even if successful, such agreements may not be able to fully accommodate the dynamic nature of tax competition, the conditions and implications of which constantly change. In other words, what is good and feasible today for a group of countries may not be so a few years into the future.

One important issue discussed in Keen and Konrad (2012) is whether tax coordination across a subset of countries is feasible and worth undertaking. The question is relevant for the simple reason that it is virtually impossible for countries to coordinate in the absence of a supra-national entity that can enforce commitment (and, if need be, arrange compensating payments) – because some countries may gain more from opting out. And since such supra-national structures exist only regionally (e.g. in the EU), in practice the question of tax coordination is often one among a sub-set of countries rather than worldwide. The literature suggests, and the empirical literature seems to support,9 that a sub-set of countries can benefit from coordinating, when its taxes exhibit “strategic complementarity” with the rest of the world – that is, when non-participating countries’ best response is to move their tax rates in the same direction as the sub-set. The issue of regional coordination is particularly important for the EI sector, where production of certain minerals can be highly concentrated in a few countries, and the setting of minimum tax rates may help such countries extract more revenues from the resource rent – countries need not be located in the same geographical region, and given the concentration of the resource, agreement on minimum rates might be easier to reach.

Devereux and Loretz (2012) review the empirical literature on source-based corporate tax competition, one of the main forms of competition that we are concerned with in this chapter.10 Countries can compete on many elements of the corporate tax: income shifting is usually sensitive to differences in statutory tax rates, while real investment is sensitive to effective tax rates.11 The authors review 74 studies of tax competition, classified according to the complexity of the methodology (trends, domestic determinants, and strategic interactions12) and the studied variables (statutory indicators, tax revenues, other taxes, and expenditures). They conclude that downward corporate tax competition has taken place in the EU and that accession of new (smaller) member states has exerted further pressure on EU CIT rates.13

Genschel and Schwarz (2011) provide a non-technical account of the literature, covering issues related to competition and coordination in consumption, labor, and corporate taxation. Their survey of empirical studies concludes that international tax competition has been ongoing mostly in the area of corporate taxation and to a lesser extent in the area of excise taxation. The revenue loss attributed to this last form is insignificant given that it occurs primarily in populated areas located close to international borders, and its impact declines rapidly with distance from the borders. There is no evidence that countries compete in the areas of value-added and personal income taxes. These results are broadly consistent with the empirical literature on the sensitivity of investment, labor, and consumption to cross-country differences in tax rates. Interestingly, the authors predict that CIT competition will slow down, primarily because of the impact of the 2008 crisis on corporate profits and revenue, which has constrained government budgets, and because the international community is more willing to address certain forms of tax competition through tax coordination – the authors refer primarily to the BEPS initiative (OECD 2013a, 2013b) and its potential impact on profit shifting.

Another strand of the literature on tax competition concerns that between the various levels of government within a single country, such as federations and unitary nations with regional governments. The insights from this literature, reviewed by Boadway and Tremblay (2012),14 may have implications for competition and coordination among countries. One element of interest is “vertical tax competition and coordination” and the related issue of “revenue assignment” between the central and regional governments. The implications of vertical tax competition for countries rich in natural resources are potentially important for two reasons. First, many of these countries, including those in the developing world, have regional governments that exert significant political influence and to which some fiscal powers have been devolved by their constitutions (e.g. Algeria, Mali, Niger, Democratic Republic of Congo, Indonesia). Regions often regard the resource as belonging to them (even when it is legally owned by a unitary state), proceeds from which are to be spent as they see fit. There are tensions between this view and the objectives of a central government that may be more concerned with issues that are beneficial to the country as a whole, such as equity, education to all citizens, health services for those who cannot afford it, and so forth, all of which need to be financed (partially at least) from the resource rent. The higher the share of resource revenue in total tax revenue, the higher are these tensions and the more complicated is coordination between the center and the regions.15 In low-income countries, these issues are compounded by the fact that mobilization of non-resource tax revenue still faces many hurdles, especially those taxes considered best suited to serve equity objectives, such as personal income taxes and property taxes (IMF, 2011).

Second, absent a tax coordination framework at the national level, it will be difficult for a federated country to coordinate its tax policies effectively with other countries. This is because local tax setting can undermine (and render non-credible) the coordination efforts of a central government when both levels of governments share the same or similar tax bases. In federated developing countries, the standard coordination model that has so far prevailed is that of the central government raising most tax revenue and then transferring a portion to state governments. In principle, this model makes sense since it obviates the need to coordinate tax policies within a federated structure, but it has proven untenable, especially in federations rich in natural resources – for the political reasons noted already. A combination of both transfers and tax assignment is likely to be more sustainable over time, as experience has shown in Canada, Germany, and the United States.

Despite the fact that corporate tax competition has been ongoing, corporate tax revenue as a share of GDP has not declined and has even increased in some countries (Devereux and Sørensen, 2006; Norregaard and Khan, 2007; Stewart and Webb, 2006).16 A number of explanations have been provided for the case of OECD countries, including base broadening (Devereux and Sørensen, 2006); increase in incorporation as a vehicle of earning business income in a sample of EU countries, or, put differently, a shift from personal income taxes to CIT (de Mooij and Nicodème, 2006), reflecting growing differences in top CIT and PIT rates; privatization, or a shift from dividend income paid by state-controlled companies to governments to CIT (Azmat, Manning and Van Reenen, 2007), and high profitability of certain sectors, particularly the financial sector (prior to the 2008 crisis). There is also the argument that reductions in CIT rates may have had Laffer-type effects, leading to large increases in investment and earnings; but Devereux, Griffith and Klemm (2002) report that marginal effective tax rates have generally not fallen greatly, casting doubt on the Laffer-effect hypothesis.17

Keen and Mansour (2010b) report similar results for the robustness of CIT revenue in Sub-Saharan Africa, despite (and unlike OECD countries) base narrowing.18 They suggest that the increase in the profit-to-GDP ratio, due, among other things, to growth in new sectors such as mobile telephony, may have played a role in preserving CIT revenue. However, more recent evidence suggests that CIT rate competition in the WAEMU is associated with a decline in CIT revenue for some member states (Mansour and Rota-Graziosi, 2013).

In summary, tax competition is happening, mostly in the CIT area, but revenue from this source has so far been broadly resilient. Whether this will continue is not clear, especially given the 2008 crisis and its impact on corporate profits and CIT revenues, which is likely to cause shifts in the worldwide allocation of capital and may prompt governments to take policy measures that they would not have considered before.19 It seems though that effective tax coordination at an international level is unlikely to occur in the foreseeable future, and so tax competition and its consequences will remain an issue on governments’ policy agendas. Even regional tax coordination, which could yield some benefits, may be difficult to achieve – as the experience of the EU in promoting the CCCBT (discussed in Chapter 2, this volume), WAEMU, and the ongoing work on a code of conduct in the East African Community (EAC) suggest.

3 Tax competition in extractive industries

Is tax competition in EIs happening? What can countries do about it? And what are the difficulties that countries may face in considering some forms of coordination?

3.1 Do countries tax-compete in extractive industries?

The question of whether countries tax-compete in EIs is relatively easy to answer if we limit it to the CIT area, since as we just discussed this is happening in most OECD countries and many developing countries. But it is important to note that the decline in CIT rates has been different across countries and that there is evidence suggesting that those rich in natural resources have reduced their rates less than those that are not. In Sub-Saharan Africa, for example (Figure 13.2), the decline in the average CIT rate between 1990 and 2010 was less in resource countries (10.3 points)20 than in non-resource countries (13.7 points) and much less in countries very rich in natural resources such as the Central African Economic and Monetary Community (CEMAC) region (3 points only).

Figure 13.2CIT rates in Sub-Saharan Africa

Beyond CIT competition, the analysis is complicated because standard CIT provisions, especially those related to the tax base, do not always apply, and the EI sector is generally subject to other taxes (e.g. royalties) or non-tax contractual arrangements. This is particularly the case for oil and gas, where many countries, especially developing countries, use production sharing agreements to extract their share of the resource rent. The mining sector is also subject to special rules in many developing countries by which governments sign contractual agreements with foreign investors for specific (typically ring-fenced) mining projects. These special tax characteristics of mining taxation systems make tax-competitive behavior difficult to identify. Different state participation schemes, in fiscal terms that in substantive terms are no different from taxation, further blur the picture and complicate the assessment of tax competition phenomena in EIs.

There is anecdotal evidence that policy makers consider other countries’ EI tax regimes when setting tax policy. For example, IMF technical assistance in resource taxation policy often benchmarks a country’s tax system as it applies to EIs against other countries’ at the request of countries’ authorities, concerned with the competitiveness of their fiscal regimes. Note, however, that the emphasis of policy makers is often on CIT and royalty rates; the tax base is rarely a concern – or not as much as it should be. This type of anecdotal evidence says little about the form of tax competition or its direction; a country may be concerned with upward tax competition21 and want to compare to other countries in order to gauge by how much it can increase its tax rates. There is some evidence from IMF technical assistance that this has occurred since the late 1990s as commodity prices soared to unexpected levels – although governments’ concerns may have shifted since commodity prices started declining in 2014.

A more robust way to weigh whether countries tax-compete for locational decisions in EIs is to examine average effective tax rates (AETRs). As we noted earlier, the AETR is the best available measure to gauge the incentives to locate in one jurisdiction among a number that offer similar investment opportunities. We estimate these for a sample of copper producing countries and a sample of gas producing countries at two different times: early 2000s and early 2010s. During this decade, production and prices rose dramatically worldwide: copper prices almost quadrupled, while production increased by 25 percent only (Figure 13.3); gas production increased by 40 percent while price almost tripled (Figure 13.4), except in North America, where domestic natural gas is abundant (including shale gas).22

Figure 13.3Worldwide mine production and prices of copper, 2000–2012

Source: International Copper Study Group (ICSG).

Note: Prices are annual averages (in USD) from the London Metal Exchange.

Figure 13.4Worldwide production and prices of natural gas, 2000–2012

Source: BP Statistical Review of World Energy, June 2013.

Note: Bcf/day means billion cubic feet per day; prices are in USD per million British thermal units; for Europe unweighted average of UK and German prices is used.

Figure 13.5 shows the AETR23 results for a hypothetical copper mining project.24 The rates increased in six countries and decreased in Canada (British Columbia) and India, where they were relatively high in the early 2000s – the small decline in the Brazil AETR is due to a more favorable depreciation scheme. Moreover, there is some convergence of AETRs. These results are mainly driven by an increase in royalty rates25 between the two periods in all countries in the sample. CIT rates remained constant in all countries for the past decade, with the exception of Canada, where the decline in the AETR is mainly due to a decline in the federal government CIT rate from 22.12 to 15 percent, and India, where the CIT surtax was reduced from 10 to 5 percent. Since all types of ferrous metals are generally subject to the same CIT and royalty regimes, an extension of the analysis to other minerals will likely yield similar results.

Figure 13.5AETRs for copper mining in selected countries

Source: Authors’ calculations.

BC: British Columbia.

Figure 13.6 shows the AETR26 for a stylized gas project.27 Rates increased in most of the countries, except Malaysia, Norway, and Nigeria, where they remained broadly constant, and Canada (British Columbia) where they decreased. As in mining, convergence of AETRs is easy to observe. The case of Tanzania and Mozambique may be explained by massive offshore natural gas discoveries – hence, improved geological prospectivity. The same is true for Poland, where vast resources of shale gas were discovered, prompting the government to design a new regime. In neither of these can the increases be explained by the CIT. In Indonesia, Tanzania, and Mozambique, they stem mostly from changes to their production sharing formulae. In the United Kingdom and Poland, it is a result of an additional tax on cash flow.

Figure 13.6AETRs for natural gas production in selected countries

Source: Authors’ calculations.

The increase in the AETRs in many countries suggests that downward tax competition in the sector may not be a concern. Indeed, it is upward tax competition that seems to be a concern, which may be related to the issue of time inconsistency or, as noted earlier, the prominent role of foreign ownership of large resource projects, especially in developing countries. In this case, one form of coordination that countries may want to pursue is to impose maximum tax rates. Maximum rates would allow countries not to pass the point at which tax levels could hurt optimal production, hence revenues. They are also a form of signaling to investors that tax rates would not reach confiscatory levels. But at what level should a maximum tax rate be set? Figure 13.6, for example, suggests a maximum rate close to 100 percent, which may not be useful as a signaling mechanism. This is an issue to which no clear-cut answer can yet be given.

3.2 Is tax coordination in extractive industries needed?

This section considers whether tax coordination in EIs is (or is not) needed, taking into account the general discussion on theory and empirics and the AETRs presented in the previous section.

The main argument in favor of not coordinating tax policy for EIs relates to the presence of location-specific rent. Theory suggests that countries can tax at 100 percent the return that exceeds the investors’ required rate of return (and other compensation for managerial and technical expertise) when the excess is due to such rent. The intuition is simple: investors make their decisions based on the risk-adjusted post-tax rate of return; when such a hurdle is reached, they are insensitive to the level of the tax rate because the resource is immobile. Provided that such a hurdle return can be observed or calculated, a country can impose tax at a rate of 100 percent once an investment project has realized its required rate of return – so long as there are no alternatives elsewhere yielding a higher net return. The choice of the tax structure is, however, key to the validity of the argument; it has to be perfectly neutral on investment decisions – as with, for instance, a cash-flow tax with remuneration (at an appropriate rate) or refund of negative cash flows or an allowance for corporate equity.

The theoretical argument has found some support in the empirical literature. For example, in his study on FDI in the EU, Stöwhase (2005) estimates the tax elasticity of outward FDI from three major EU countries by sector;28 he finds that FDI in the primary sector is insensitive to differences in both forward-looking and backward-looking AETRs across countries. These results are important because empirical studies about the sensitivity of FDI to tax-rate differentials tend to find significant and relatively high elasticities on average, without differentiating the elasticity by sector.29

Boadway and Keen (2010) develop a number of other arguments why tax competition in EIs might occur and whether tax coordination is desirable. These arguments are not strictly related to the extraction of the resource itself but more to the market conditions of needed capital and labor inputs and strategic considerations by large firms. We review them here, and add a few of our own.

The first argument is that it is so hard to attract certain special skills (managerial and technical) or highly specialized equipment in EIs that are short in supply that governments must provide an overall attractive tax package. There are two difficulties with this argument. First, it is not clear why the tax incentive must be provided to the project rather than directly to the skill that is in short supply. From a cost-effectiveness point of view, it makes more sense to compete on the skills needed by, for example, providing wage-based incentives rather than on the project as a whole by, for example, providing time-bound CIT holidays. Second, basic supply-and-demand theory tells us that if something is in short supply, its value increases enough to encourage new entrants. The argument is therefore one of transition to a new equilibrium – although transition in EIs could take some time given the nature of the industry and commodity price volatility.

Another argument is that under certain market conditions, a large firm with influence on the international price of a product may deliberately choose to restrict its production to increase prices or keep them from falling. It is not clear what the tax competition and coordination implications of such market imperfection are. Might other countries provide tax incentives in order to compensate for the shortage? It is not clear what role tax coordination can play in order to discourage firms from constraining production. In fact, constraining countries from tax competing could play in favor of the firm constraining demand and make the situation worse for all countries – in terms of paying higher consumption prices, but the country where production is constrained presumably gains higher tax revenues given higher prices and hence has no incentive to coordinate.

The third argument is that countries may simply want to behave like others for marketing purposes. A country may perceive it to be beneficial when it can say that its tax regime is similar to neighboring countries or to the largest producers of a given resource commodity. For example, if a country has a much higher AETR than other countries, investors may shun it for profitability reasons; on the other hand, if the AETR is too low, investors may take it as an indication of poor governance or lack of credibility (i.e. that once the resource is discovered and reserves estimated, the country would increase its taxes). The AETR analysis in the previous section lends some support to this; rates for copper and natural gas converged over the past decade. But convergence to a positive and higher AETR argues in favor of a maximum tax rate, not a minimum – as the standard tax competition model suggests.

There are other reasons tax coordination may be desirable. One is that politicians generally discount the future at high rates and prefer to have revenue to spend today rather than when they are no longer in power. At constant prices, this simply means encouraging firms to produce more. It would generate more revenue to the government if the value of incentives provided (in terms of revenue loss) is lower than the increase in the government’s share of revenue from additional production.

A country may also tax compete in a disputed area (with other countries) in order to make it attractive for companies to invest given the legal and, in some cases, high security risks regarding disputed areas – and, perhaps, the other non-tax factors that are typically weak in such areas, such as lack of investment in public infrastructure. Some investors are prepared to take such risks, but they also face the reputational risk – of accepting to operate in areas without clarity as to which country has a legitimate claim to the areas. Tax competition may signal in this case the inability of a government to resolve its dispute with neighboring countries – perhaps even risk of future conflict – and may discourage (rather than encourage) investors who value reputational risks.

Finally, it is possible that countries tax compete simply because they are weak in negotiating skills with major players in EIs. This is particularly the case in developing countries, where EI projects may be negotiated between investors and the government or a state enterprise, and a contract is agreed that contains the tax provisions. This situation can be aggravated by institutional fragmentation and weakness within the country and by corruption.

It is very difficult to empirically substantiate these arguments, both individually and when they interact. Some of them are driven by market imperfections, either for the resource itself or for capital and labor inputs needed to extract it; others are more closely related to the political economy of managing natural resources. In the first instance, addressing the imperfections at the country level may be a better policy response than seeking tax coordination with other countries. In all cases, however, there seems to be some support, in principle, for a minimum level of taxation, at least to protect weak countries that are new in EIs from making policy mistakes they will regret later. Note too that countries that can maintain a fairly high level of taxes should also benefit from such a strategy or at least have nothing to lose. Our analysis also suggests that a maximum level of taxation may be useful – countries may gain from signaling collectively their willingness to address the commitment problem. The question then is how this can be done. What taxes should governments coordinate? In the next section, we discuss the difficulties in coordinating taxes on EIs, which stems primarily from the variety of policy instruments used by governments to extract their share of the resource rent.

3.3 Difficulties in coordinating taxes on extractive industries

Tax coordination in EIs is made difficult by a number of factors that are specific to the sector or perhaps more common and important than in other sectors.

First, the number of taxes or levies deployed to tax EIs is typically higher than in non-EI sectors, and these taxes are more complex in their application than those in other sectors. This is particularly the case in developing countries, where in addition to profit taxes and royalties, governments may have a stake in EI projects (either free or carried), may require signature and production bonuses, and may create state enterprises that play a key role in research and development as well as extraction.30

The number of taxes is important because coordination of one tax could shift tax competition to other taxes, with non-trivial consequences for revenue and efficiency.31 For example, if coordination imposes a minimum CIT rate and a common base, countries can cut their royalty or rent tax rates to compete. In this case, the objective of securing revenue by imposing a minimum rate is undermined by the loss of royalty revenue; moreover, the temporal profile of revenue changes. The reverse holds if coordination imposes a maximum CIT rate; a country constrained by the maximum and seeking to preserve revenue may increase export taxes, royalties, or other fees, which are more distortionary than profit taxes. Similar arguments apply to competition through tax bases rather than rates. Table 13.1 lists the main taxes that are applied to EIs around the world and provides a very brief indication of possible variations across countries in the calculation of the base. It is easy to see that for coordination to have a chance of succeeding, in terms of imposing a hard constraint on countries’ EI tax policies, rationalization of the number of taxes is a prerequisite – coordination on all of these taxes would be impossible to achieve.

Table 13.1Taxes generally applicable to extractive industries
Input taxesOutput taxesProfit and rent taxes
External tariffs: Prevalent mostly in developing countries and applied to both intermediate and capital inputs.Export tariffs: Rare in general but exist in some countries, especially on raw minerals. They function as royalties in some cases (more convenient to levy at the point of export).
Excise taxes: Specific or ad-valorem. Can be very important, especially on energy products used in processing minerals, including electricity generation in remote areas.
Equity participation (free) and production sharing: Latter varies in the definition of the base, especially in terms of limitations on deductibility of certain expenses (much like the standard profit tax).
Royalties: Can be specific or (more common) ad-valorem. Latter often calculated on a base that is somewhere between turnover and accounting profits (measured on accrual basis).
Rent taxes: Surtaxes, progressive profit tax, various types of cash-flow taxes.
Sales taxes: Can fall on inputs to the extent not refunded. Neutral if VAT-style with proper refund or suspension mechanism.
Standard profit taxes: Differences in base across countries are mainly in treatment of capital goods, loss carry-forward, and limitations on certain expenses (e.g. interest expense; HQ management fees, etc.)
Signature bonuses: Vary enormously across countries (some do not have them). Could also be classified as output taxes.Production bonuses: Vary with the level of production and may act as imperfect rent taxes.
Surface fees: Taxes on land as an input. Generally not very important in terms of revenue generation.
Source: Prepared by the authors.
Source: Prepared by the authors.

Second, production sharing agreements (PSAs) are widely used in developing countries for capturing resource rent. These are contractual arrangements whose fiscal terms, often negotiated between investors and government, are locked for fairly long periods and often with guarantees of stability.32 If stability is asymmetric, providing guarantees against increases in taxes but allowing for decreases, coordination that attempts to increase taxes or stem a decline in tax rates may be ineffective for existing investment.

Third, the abundance of resource revenue may weaken the incentives for good tax policy practice, including tax coordination. If the gains from coordination are marginal relative to existing natural resource wealth, a country may simply prefer to set policy freely rather than be constrained. There are exceptions, however, such as the case of joint development zones (e.g. the Saudi-Kuwaiti neutral zone), where two or more countries may be more amenable to agree on a common tax regime to allow a company to extract minerals from the zone and share the revenue, as discussed in Chapter 10 (Cameron, this volume) and Chapter 11 (Daniel, Veung and Watson, this volume). It is fair to say though that non-tax considerations such as those noted earlier play a more important role in these situations.

4 Experience in tax coordination in extractive industries

Experiences in tax coordination are generally not sector specific; the only exception we know of is the WAEMU Mining Code regulation of 2002. In this section, we review experiences in direct and indirect tax coordination, with an emphasis on the implications for EIs.

4.1 Direct tax coordination

Many of the regional economic blocs around the world have reached various levels of integration of their trade regimes, but only the EU and WAEMU have reached some modest degree of coordination in their direct taxes, primarily certain aspects of the corporate income tax and the taxation of portfolio income, and only WAEMU has attempted to coordinate the taxation of the mining sector.

There are two separate measures in the EU that aim to tackle tax competition: the Code of Conduct for business taxation and the State Aid Rules; both are general and not limited to direct taxes. A third initiative currently being debated, and strictly related to the corporate tax, is the common consolidated corporate tax base (CCCTB).33 All the measures are of general application, and none of them focuses specifically on EIs.34 We consider each in turn.

The Code of Conduct for business taxation was one of the first proposed explicit tax coordination measures.35 It has never become a binding law – it is merely a political commitment of the member states, endorsed by a resolution of the European Council,36 to respect principles of fair competition and to refrain from harmful tax competition. The code covers both tax laws or regulations and administrative practices (e.g. concessionary tax decisions and rulings) and refers to any tax measure that may provide for a significantly lower effective level of taxation than that generally applied in a given member country. Only tax measures used to support the economic development of particular regions within the EU received special attention. The Code does not consider them to be potentially harmful, as long as they are in proportion to “the aims sought”, which is difficult to achieve in tax design, particularly for measures with no limits on benefits – such as tax holidays.

State Aid rules, contained in the EU treaty, may be considered another form of tax coordination as they contribute to the objective of unimpeded functioning of the internal market by tackling harmful tax competition. They do not refer explicitly to tax policy design. Any measure that distorts or threatens to distort competition by favoring certain undertakings or the production of certain goods and thus affecting trade between the member states may be considered a state aid. It is then straightforward that certain tax measures that may provide recipients an advantage by reducing or relieving them of taxes of general application (e.g. accelerated depreciation, provisions, etc.) are potentially liable to these rules.37 In contrast to the Code of Conduct principles, the member states are legally obliged to comply with the State Aid rules and must seek a prior approval from the European Commission before granting tax subsidies. The European Court of Justice, with a power to revoke discriminatory tax features, plays an important role in enforcing these rules.

The proposed CCCTB goes much farther than the two preceding measures in relation to the taxation of corporate income. The proposal is to create common rules for calculating the corporate income tax base and allocating it to member states according to a formula. It is important to note, however, that tax rates would continue to be set at the national level. That would imply a significant degree of tax coordination with respect to the use of tax allowances, deductions, including depreciation schemes, loss carry-forward, and other tax measures affecting the calculation of the CIT base. In effect, corporate tax competition under a CCCTB would be driven primarily by the statutory rate – since base competition would be neutralized. However, the CCCTB, as proposed in 2011, is intended to remain voluntary;38 it would exist as a separate tax code, in parallel to national ones. In that sense, its coordinative impact may therefore be limited.

The second experience with direct tax coordination is the WAEMU.39 Unlike the EU, the WAEMU, since 2008, coordinates the CIT base and rate not through a Code of Conduct but through legally binding directives: one that defines the CIT base and one that limits the rate to a minimum of 25 percent and a maximum of 30.40 Coordination of the base leaves some flexibility for countries to compete. For example, the rules for tax depreciation and limitations on interest expenses are left for national laws to set. But the most important source of tax competition among WAEMU countries is through the derogatory regimes provided for in non-tax legislation, such as Investment Codes, Free Zone Codes, and other sectoral codes, which are explicitly permitted under Article 8 of the CIT base directive.41 These are the main reasons tax coordination among WAEMU states has not been effective. Indeed, Mansour and Rota-Graziosi (2013) argued that the CIT directives may have intensified tax competition by limiting member countries’ abilities to act within their tax laws but giving them unfettered liberty to legislate tax provisions in non-tax laws.

Another peculiarity of the WAEMU, and the only example we know of, is a regulation42 specifically coordinating taxation of the mining sector, introduced in 2003. It provides that firms are subject to the general tax laws of member states and to a royalty whose base and rates will be determined later by regulation – to our knowledge, these have not been issued. The regulation also provided for stability of the tax regime – presumably, both for taxes imposed at the national level and the royalty that was to be fixed regionally for all member states – during the life of the investment. The stability is asymmetric, insuring against increases in taxes but allowing taxpayers to benefit from reductions.43 In terms of tax incentives, the regulation provided for the exemption of virtually all taxes and fees during the exploration phase. The main incentives provided during the production phase are accelerated depreciation and a three-year tax holiday from profit and payroll taxes; the modalities for the coverage and calculation of accelerated depreciation have not been issued.

Together with the CIT directives, the mining regulation would have harmonized substantially member states’ tax regimes and limited their freedom to set tax policy for the sector. However, the absence of agreement on royalty rates in regulations and the fact that countries can tax-compete for CIT through sectoral legislation make coordination ineffective. All this is indicative of the political difficulties of tax coordination in general and in the EI sector more specifically.

4.2 Tariff and indirect tax coordination

In this sub-section, we examine briefly the implications of tariff and indirect tax coordination as it affects the EIs, drawing from IMF technical assistance experience in Sub-Saharan Africa. As with direct tax coordination discussed in the previous sub-section, measures of tariff and indirect tax coordination around the world are generally not sector specific. But because they deal with the taxation of goods and services rather than incomes, they can be designed to mimic, to a significant degree, turnover taxes. For example, an excise tax can be designed to mimic a royalty or an export tax in a country that produces gold primarily for export.

4.2.1 Tariffs

In principle, reducing import tariffs is desirable since it enhances production efficiency and reduces consumer prices. Provided that countries are not constrained by the choice of tax instruments and can substitute efficient taxes (e.g. VAT, rent tax) for distortionary tariffs (and this is a strong assumption, particularly for low-income countries), tariff competition should improve welfare without reducing government revenue.44 Importantly, countries can achieve such results unilaterally, without any need for coordinating their actions. This is particularly important for resource-rich developing countries, where capital goods for EIs are generally imported; replacing tariffs on capital goods with more efficient domestic taxes (e.g. property taxes or consumption taxes) can enhance a country’s attractiveness as a destination for EI investment.

This is broadly what happened in developed countries in the past 50 years: tariffs and cascading sales taxes have been replaced by the VAT, which has become a major revenue source. The story in developing countries is somewhat different. These countries seem to be constrained in the choice of policy instruments to raise revenue; some have not been able to fully replace the revenue loss from tariff reduction with other revenue sources, such as the VAT. Precisely why this is the case is not clear. There is no empirical evidence on why some countries have done better than others in replacing lost trade revenue.

After a marked decline in the 1990s, the share of tariff revenues in GDP in low- and lower-middle-income developing countries has stabilized during the past decade – at around 2.5 percent for the first group and 4.5 percent for the second. Tax systems have become increasingly reliant on the taxation of final consumption goods and less on that of inputs and capital goods. One important reason for this is that tariff coordination in a number of regions (e.g. WAEMU, EAC, Common Market for Eastern and Southern Africa [COMESA]) was used as an opportunity to reduce the level and number of tariffs on inputs and capital goods – EAC and COMESA have eliminated the latter. So even if not needed, tariff coordination through the formation of custom unions has given countries the policy momentum to reduce tariff rates on inputs.

Another important reason is the interaction of tariff and domestic tax policies. As noted earlier in this chapter, domestic tax policy is fraught with tax incentives legislated in non-tax laws. These laws, including EI-specific laws, typically provide time-bound exemptions or reduction in tariffs which result, de facto, in the application of effective tariff rates that are more numerous and generally lower than statutory rates. Tariff coordination through customs unions obviates the need to provide tariff exemptions and renders the tax system more transparent and visible to investors.

4.2.2 Value-added taxes

The VAT is an area where misconceptions about whether coordination is desirable abound. Being a consumption tax, the VAT should, if well designed, not be subject to the forces of tax competition – at least not those affecting investment location decisions, particularly in the EI sector – and, as noted earlier, empirical evidence tends to support this. One issue, though, that deserves attention in developing countries, is the implication of the limitation on refunds for the cost of capital and hence for tax competition and coordination. This is an area in which coordination could have a similar impact as that of VAT rate convergence in WAEMU.45 For EIs, the issue arises primarily in relation to intermediate inputs, such as energy consumption and capital goods, though the latter are frequently exempted during the development phase of a project. The issue is not merely a policy issue; administrative practices frequently constitute a significant constraint for EI firms in receiving their VAT refunds within reasonable time (see Harrison and Krelove, 2005).

Delaying or disallowing refund payments can be seen as upward tax competition. This can be particularly important when governments are constrained by stability clauses not to impose new taxes on EIs or to increase existing ones. In such a case, VAT coordination, especially rules and regulations relating to the payment of refunds, can be seen as a mechanism to alleviate, collectively, the commitment problem and is beneficial from an efficiency point of view – since it reduces effective tax rates on inputs.

Difficulties in the payment of VAT refunds often give rise to exemptions and other types of reliefs. Rather than seeking a solution to the underlying problem with timely refunds (or suspension), countries often resort to granting concessions. Such a practice quickly proliferates. Investors, while lobbying for input VAT concessions, use examples of other countries to win similar favorable treatment. For governments, it can be easier (and more visible) to create a level playing field by granting exemptions in line with other countries rather than refunding.

There are a number of other specific challenges that VAT may pose to businesses operating in EIs. Ambiguous treatment of settlements between partners in unincorporated joint ventures (not uncommon in the petroleum industry), different approaches to the taxation of services supplied to offshore facilities, or impediments to zero-rating of exports46 are just a few examples (see Świstak, 2016). They create uncertainty and add to compliance costs and hence may affect investment decisions. Pressures to alleviate unrecovered or inappropriate input taxes are best met with ensuring proper design and implementation, rather than by ad hoc exceptional treatments.

4.2.3 Excises

Excise taxes affect the production sector of the economy primarily when they are imposed on inputs. For EIs, energy products (e.g. diesel, electricity) tend to be a significant input. Therefore, countries may compete downward on excises in order to attract particular elements of the extraction process. For example, an EI firm may decide to extract a mineral (say copper or cobalt) in one country but do most of the work needed to bring it to a concentrate form in a neighboring country. This would call for coordination of excises on energy products among neighboring countries, such as agreeing on minimum rates and common rules to value the tax base.47 But since many non-tax factors come into play in making such decisions, and since energy products are more widely consumed (and not just by EI firms), it is unlikely that the EI sector on its own could be the main motivator for coordination.

5 Concluding remarks

The theoretical literature on tax competition and coordination does not provide unambiguous answers to the questions on tax competition and coordination posed in this chapter. The argument that downward tax competition in EIs may be causing significant revenue loss does not seem to hold. New evidence presented here suggests that AETRs in key EIs have actually increased during the first decade of this century, a period characterized by significant increases in commodity prices, and there is some evidence that CIT rates in resource-rich countries have declined much less than in non–resource-rich countries.48 This suggests that tax competition in EIs may be less of a concern to governments than it is in other sectors.

It can, of course, be dangerous to extrapolate from past trends, and the analysis suggests that there can be a case, for a set of countries producing a homogenous resource, to impose minimum and maximum tax rates on EIs – hedging against the time inconsistency issue and providing some assurances that tax rates will not be increased to confiscatory levels once substantial costs have been sunk. Given the complexity of tax systems in EIs, in particular the number of taxes deployed by governments in developing countries, it is by no means trivial how such coordination can be accomplished.

Our analysis does not consider situations in which the tax treatment of EIs is negotiated between government and investors and the terms are not publicly available – although this is implicit, we use only tax rates and base rules that are publicly available. In other words, tax competition may be taking place in ways that are difficult or impossible to observe and measure. This is quite possible given that transparency and political economy issues play significant roles in EIs. An interesting question then is whether tax coordination is realistically possible in such situations.

Tariff and indirect tax competition, when they affect production decisions, can improve incentives and efficiency and can be replaced by less distortionary taxes to preserve revenues. Customs unions may create an impetus to lower tariff rates and replace cascading sales taxes with VATs, but they are by no means necessary since countries could achieve the efficiency benefits of such policies without coordination and without loss of revenues. Getting national policies right in this area seems to be the overriding factor for improving the conditions for EI growth – especially in terms of reducing the cost of capital for exploration and development.

References

    AzmatGhazalaAlanManning and John VanReenen. (2007) “Privatization, Entry Regulation and the Decline of the Labor’s Share of GDP: A Cross-Country Analysis of the Network IndustriesDiscussion Paper 806 (London: London School of Economics and Political Science, Centre for Economic Performance).

    BaunsgaardThomas and MichaelKeen. (2010) “Tax Revenue and (or?) Trade LiberalizationJournal of Public Economics94(9–10) 563577.

    BoadwayRobin and MichaelKeen. (2010) “Theoretical Perspectives on Resource Tax Design” in PhilipDanielMichaelKeen and CharlesMcPherson (eds) The Taxation of Petroleum and Minerals: Principles Problems and Practice (London, New York: Routledge) 1374.

    BoadwayRobin and Jean-FrançoisTremblay. (2012) “Reassessment of the Tiebout ModelJournal of Public Economics9610631078.

    DanielPhilip and Emil M.Sunley. (2010) “Contractual Assurances of Fiscal Stability” in PhilipDanielMichaelKeen and CharlesMcPherson (eds) The Taxation of Petroleum and Minerals: Principles Problems and Practice (London, New York: Routledge) 405424.

    DevereuxMichaelRachelGriffith and AlexanderKlemm. (2002) “Corporate Income Tax Reform and International Tax CompetitionEconomic Policy17 (35) 451495.

    DevereuxMichael and SimonLoretz. (2012) “What do We Know about Corporate Tax CompetitionWP 12/29Center for Business Taxation, Oxford University.

    DevereuxMichael and Peter BirchSørensen. (2006) “The Corporate Income Tax: International Trends and Options for Fundamental ReformsEconomic Paper 264 (Brussels: European Commission).

    de MooijRuud and SjefEverdeen. (2003) “Taxation and Foreign Direct Investment: A Synthesis of Empirical ResearchInternational Tax and Public Finance10(6) 673694.

    de MooijRuud and GaëtanNicodème. (2006) “Corporate Tax Policy, Entrepreneurship and Incorporation in the EUWorking Paper 1883 (Munich: CESifo).

    GenschelPhilipp and PeterSchwarz. (2011) “Tax Competition: A Literature ReviewSocio-Economic Review9339370.

    HarrisonGraham and RussellKrelove. (2005) “VAT Refunds: A Review of Country ExperienceWorking Paper 05/218 (Washington, DC: International Monetary Fund).

    International Copper Study Group; 2010 and 2012Annual Reports.

    International Monetary Fund. (2005) Dealing with the Revenue Consequences of Trade Reforms (Washington, DC: International Monetary Fund).

    International Monetary Fund. (2011) Revenue Mobilization in Developing Countries (Washington, DC: International Monetary Fund).

    International Monetary Fund. (2012) Fiscal Regimes for Extractive Industries: Design and Implementation (Washington, DC: International Monetary Fund).

    KeenMichael and Kai A.Konrad. (2012) “International Tax Competition and Coordination” in Alan J.AuerbachRajChettyMartinFeldstein and EmmanuelSaez (eds) Handbook of Public Economics vol. 5 (Amsterdam, Boston, Heidelberg, London, New York, Oxford, Paris, San Diego, San Francisco, Singapore, Sidney, Tokyo: North-Holland) pp. 257328.

    KeenMichael and Jenny E.Ligthart. (2002) “Co-ordinating Tariff Reduction and Domestic Tax ReformJournal of International Economics56490507.

    KeenMichael and Jenny E.Ligthart. (2005) “Co-ordinating Tariff Reduction and Domestic Tax Reform under Imperfect CompetitionReview of International Economics13385390.

    KeenMichael and MarioMansour. (2010a) “Revenue Mobilization in Sub-Saharan Africa: Challenges from Globalization I - Trade ReformDevelopment Policy Review28(5) 553571.

    KeenMichael and MarioMansour. (2010b) “Revenue Mobilization in Sub-Saharan Africa: Challenges from Globalization II - Corporate TaxationDevelopment Policy Review28(5).

    LejeuneIne. (2011) “The EU VAT Experience: What are the Lessons?The VAT Reader: What a Federal Consumption Tax Would Mean for America (Falls Church, VA: Tax Analysts).

    MansourMario. (2014) “A Tax Revenue Dataset for Sub-Saharan Africa: 1980–2010Revue d’Économie du Développement2299128.

    MansourMario and CaroleNakhle. (2016) “Fiscal Stabilization in Oil and Gas Contracts: Evidence and Implications” SP 37 (London: Oxford Institute for Energy Studies).

    MansourMario and GrégoireRota-Graziosi. (2014) “Tax Coordination and Competition in the West African Economic and Monetary UnionTax Notes International74 (2) 183204.

    OECD. (2013a) Addressing Base Erosion and Profit Shifting (Paris: OECD).

    OECD. (2013b) Action Plan on Base Erosion and Profit Shifting (Paris: OECD).

    OsterlohSteffen and FriedrichHeinemann. (2013) “The Political Economy of Corporate Harmonization - Why European Politicians (Dis)like Minimum Tax Rates?European Journal of Political Economy291837.

    NorregaardJohn and TehminaKhan. (2007) “Tax Policy: Recent Trends and Coming ChallengesWorking Paper 07/274 (Washington, DC: International Monetary Fund).

    StewartKenneth and MichaelWebb. (2006) “International Competition in Corporate Taxation: Evidence from the OECD Time SeriesEconomic Policy45153201.

    StöwhaseSven. (2005) “Tax-Rate Differentials and Sector-Specific Foreign Direct Investment: Empirical Evidence from the EUFinanzArchiv61 (4) 535558.

    ŚwistakArtur. “Value Added Tax in Extractive IndustriesWorking Paper IMF.

Notes

We thank Michael Keen, Victor Thuronyi, and Philip Daniel for very helpful comments and Victor Kitange and Chandara Veung for collating comparative tax data and undertaking effective tax rates analysis.

The word “collectively” implies that, individually, some countries might lose and others might gain from tax competition. This also implies that for tax coordination to work, and hence for revenues to be higher relative to uncooperative tax setting in all jurisdictions, countries that gain from tax competition must be compensated; the compensation provides the incentive to cooperate.

Downward tax competition occurs when countries bid down their tax rates in an effort to attract investment they believe would otherwise locate in other countries.

There is an ongoing and old debate in the literature regarding whether certain levies on EIs are “taxes” or “fees” for the use of property (e.g. royalties). In this chapter, we simply consider all levies on EIs to be taxes, to the extent that they are compulsory payments to government, and their revenue and behavioral implications can be analyzed like taxes.

The G20 has recently put this form of tax competition on the top of its agenda, in the form of the base erosion and profit shifting (BEPS) initiative, spearheaded by the OECD. The initiative, very broadly speaking, seeks to tighten the design of certain elements of the OECD bilateral tax treaty model and transfer pricing guidelines and widen the scope and depth of tax information sharing across countries; the outcomes are summarized in the appendix to Chapter 2, Keen and Mullins (this volume). The OECD has sought to broaden participation in this initiative to non-G20/OECD members.

There are exceptions, however, as in some countries of the Gulf Cooperation Council; but the history of these examples suggest that the origin of low or no taxation of corporate profits in the non-resource sector is not tax competition for paper profits. Generally, countries that can rely entirely on natural resources may become unintentionally attractive tax havens because they do not need taxes, and particularly the corporate tax.

See on this Baunsgaard and Keen (2010), Keen and Mansour (2010a) for Sub-Saharan Africa, and IMF (2005).

Tax competition can, in principle, be beneficial in preventing governments from setting taxes, and hence spending, too high (see for instance the discussion in Edwards and Keen (1996) – though fiscal rules seem likely to be a preferable way of achieving this, since they do not directly restrict instrument choice.

This is because raising the tax rate set by a low-tax country leads others to set higher rates than they otherwise would, conveying a benefit to the low-tax country that more than offsets the direct impact of a small increase in its own rate. But even a minimum tax is not easy to build consensus around (see Osterloh and Heinemann, 2013).

See the review by Devereux and Loretz (2012).

This refers to competition in the taxation of corporate profits sourced in a tax jurisdiction, as opposed to residence-based taxation, which seeks to attract residency of investors rather than their investment activities. Source-based tax competition is possible because taxation in the residence country is now largely absent or (primarily in the United States) allows indefinite deferral of taxation of profits sourced in foreign countries. The basic theoretical model of this type of competition predicts (under rather strict conditions) that source taxes on corporate profits should be set to zero. Although this is largely inconsistent with what we observe in practice, the extensive provision of CIT holidays in developing countries (often renewable and targeting FDI) lends some support to this basic model.

There are two forward-looking effective tax rates commonly used in tax policy analysis: the marginal effective tax rate (METR) and the average effective tax rate (AETR). The first is a theoretical construct of the gap between the pre- and post-tax return on a unit of investment that is marginally worth undertaking. The second, also a theoretical construct in a forward-looking sense, is the ratio (in present value) of tax on expected future profits to pre-tax expected net cash flows – typically measured over the life of an investment. The METR is a measure of the incentive or disincentive that the tax system provides to investors on marginal decisions (e.g. buying new machinery, increasing exploration expenses, holding higher levels of inventory). The AETR is a measure of whether an investment project, such as mining gold in a country, is worth undertaking; it is more relevant than the METR for location decisions such as deciding to invest in one country rather than another.

The standard econometric specification to test for strategic interactions between countries is a reaction function that takes the general form ti=aΣjWijtj+bXi+ei, where i and j are country indices, t is a tax rate, w is a weight matrix used to average the ts across countries, X is a vector of socio-economic variables affecting the tax rate, and e is a normally distributed error. A positively sloped reaction function (a > 0) indicates strategic complementarities in tax rate setting between countries.

The authors are critical of previous trend analyses of rates or revenues (particularly the latter) and econometric specifications that ignore, in the explanatory variables, the studied variables in other countries. Their conclusions are primarily based on recent studies using reaction functions.

This literature on fiscal federalism is broader than the tax competition literature; it covers revenue and expenditure policy and their assignment across levels of governments.

In principle, the complexity in coordinating taxes should be related to the number of entities that need to coordinate. But experience shows that the amount of revenue at stake is also important even when the number of players is small. For example, in the Democratic Republic of Congo, copper mining is essentially concentrated in the province of Katanga, yet tax coordination between the central government and the provinces in the taxation of natural resources has proven very difficult.

This increase predates the 2008 crisis and its impact on corporate profits, which may or may not be temporary.

This evidence is becoming outdated. New research is needed, especially in light of the impact of the 2008 crisis on corporate profits and CIT revenues.

Unlike most studies, the authors construct a database that excludes from CIT revenue the share of upstream activities in mining and oil and gas, which presumably are less affected by the forces of tax competition than other sectors.

This has already happened in certain areas, such as the taxation of the financial sector in Europe.

Resource countries are defined as those that reported revenues from the resource sector (CIT on profits from upstream activities, royalties, or production sharing), regardless of their relative share in total tax revenues (Mansour, 2014).

Upward tax competition occurs when countries increase their tax rates as a response to an increase in other countries. In the EI sector, it could occur when countries decide to capture a higher share of the resource rent due to a favorable change in commodity prices or geological prospectivity (e.g. massive discoveries of natural gas in Rovuma Basin in East Africa or shale gas in Eastern Europe).

Oil and mineral prices have been falling since mid-2014. Although the recent significant drop in prices does not affect our analysis (as its time span extends only to early 2010s), we do note it may create a pressure for governments to adjust fiscal terms downward if faced with a threat of, say, mine closure. Whether countries will compete to save their mine, especially when held by a multinational company with mining operations in several countries, is yet to be seen. It will be an interesting topic for future research in the field of tax competition in EIs.

Net present value discounted at 7 percent. Income tax, royalty, and state participation, where applicable, are taken into account. No indirect taxes, withholding taxes, and surface fees are included in the calculations.

The project economics are as follows: life of 21 years, of which 3 years for exploration/development; production of 12 million tons (0.66 million annually); capital expenditures are USD 7.9 billion; decommissioning costs are USD 0.2 billion; unit operating and processing costs is USD 1.16 per ton in 2012 prices; copper price is USD 6,000 per ton, increasing at 2 percent per annum over the project’s life.

Not all royalty regimes’ bases are similar, and some have elements of net income taxation; they remain, however, generally more distortionary to marginal investment than income or rent-type taxes.

Net present value discounted at 7 percent. Royalties, profit gas, bonuses, income taxes, and additional profit taxes are taken into account. No indirect, property, and withholding taxes are included in the calculations.

Project economics are as follows: life of 38 years, of which 6 years for exploration/development and 2 for decommissioning; no condensate or other liquids are produced; production is 6 Tcf (260 Bcf annually, with a plateau reached in the sixth year of production); capital expenditures are USD 4.8 billion; decommissioning costs are USD 1.4 billion; unit operating and liquefaction cost is USD 6.1 per Mcf in 2012 prices; gas prices are USD 14 per MMBtu, increasing at 2 percent per annum over the project’s life.

The three countries are the UK, Germany, and the Netherlands. The sectors are primary (includes EIs), manufacturing, and services.

See, for example, the synthesis of empirical studies by de Mooij and Everdeen (2003).

We do not include here taxes that are levied by EI enterprises on behalf of their employees or shareholders, which include wage taxes and social security contributions, withholding taxes on dividends, interest, and various payments to non-residents. For these, the enterprise simply acts as the withholding agent for the government.

The experience of Poland may serve as an illustration of tax competition shifting to other taxes. The 2014 law on special hydrocarbon tax (Ustawa o specjalnym podatku weglowodorowym) does not provide for an extended period of loss carry-forward under the CIT (above the general limit of five years), as it could constitute a state aid (forbidden under the EU rules, see discussion in section 4.2 of this chapter). Instead, the amount of lost tax benefit resulting from non-deduction of losses due to the expiration of the five-year period may be offset against royalty payments.

For a description of and issues with fiscal stability, see Daniel and Sunley (2010) and Mansour and Nakhle (2016). The latter documents a trend toward more asymmetry in fiscal stability in the oil and gas sector since the late 1990s.

In June 2015, the European Commission presented a strategy to re-launch the common consolidated corporate tax base (CCCTB). For more details, see Communication from The Commission to the European Parliament and the Council (Brussels, COM (2015) 302 final), available at: https://ec.europa.eu/priorities/sites/beta-political/files/com_2015_302_en.pdf

Art. 100 of the EC’s Directive Proposal foresees specific rules for oil and gas in ascertaining location of sales for purposes of apportionment of the consolidated tax base, that is, place of extraction/production and not destination of sales are taken into account (Brussels, COM (2011) 12/14).

A package to tackle harmful tax competition in the European Union (1997), COM (97)564 (Annex I).

Conclusions of the ECOFIN Council Meeting on December 1, 1997, concerning taxation policy (98/C 2/01).

See EC Notice on the application of State Aid rules to measures relating to direct business taxation (98/C 384/03).

The EC is considering a proposal to make the CCCTB mandatory for certain companies. The criteria determining which companies should use the CCCTB rules are not known yet. See supra note 35.

A detailed account of the WAEMU tax and tariff coordination framework is provided by Mansour and Rota-Graziosi (2014).

These are, respectively, directive 01/2008/CM/UEMOA and directive 08/2008/CM/UEMOA.

These regimes potentially violate the State Aid rule in article 88 of the WAEMU Treaty, which explicitly prohibits “public aid that could distort competition by favoring certain enterprises or production activities” (translation from French by the authors).

Unlike a directive, a regulation must be adopted as is by member states.

There are no good reasons for a country to provide asymmetric stability. Moreover, in the WAEMU tax coordination context, stability for the CIT is of limited use since tax rates are bound by both a minimum and a maximum.

The theoretical case for this is spelled out in Keen and Ligthart (2002 and 2005).

The issue is much less important in developed countries but not insignificant. See, for example, Lejeune (2011) on the lessons from VATs in the EU. Other jurisdictions, such as the Canadian province of Quebec, limit refunds of VATs on certain inputs, including energy products.

For example, in 2013, Zambia effectively suspended zero-rating of minerals for lack of adequate proofs of export. The rules in place did not cater sufficiently for all International Commercial Terms (Incoterms rules) or situations in which minerals were exported by trading companies rather than directly by mining companies.

Transportation costs play a significant role in how competition and coordination may play out. For example, rough uncut diamonds can be transported anywhere around the world at virtually no cost relative to their market value; the case of unprocessed copper or cobalt is very different.

Our sample of countries is, however, by no means comprehensive; an extension of the analysis to a broader sample and a longer period for the AETRs may shed more light on this issue.

    Other Resources Citing This Publication