500 BEST PAID WORKERS NAMED.

FISCAL REGIME DESIGN AND REVENUE COLLECTION

RESOURCE ECONOMICS

Thought-provoking questions:

1.    What are the key concepts in resource economics?

2.    How does uncertainty in resource extraction influence the design of the fiscal regime?

3.    How might differences in resource endowments across countries result in different fiscal policies?

Good morning or good afternoon, depending where you are.

I'm privileged to have an opportunity to spend some time addressing the fiscal regimes for extractive industries. I have quite a lot of material to present and so I hope you will bear with me. I'm going to cover the fundamentals of resource economics to start with and then move on to the principles and practices of fiscal policy for extractives, then spend some time talking about fiscal regime design and some of the challenges around that and then finally move on to the question of revenue administration, which may be the last but by no means the least topic. Throughout I'll address these topics from a purely public policy perspective and assume that the countries of interest are resource rich developing countries.

The sources for what I have to say are many and the reference materials I believe can be availed only on request. This is essentially a personal synthesis of a lot of material and ideas which I hope will help you to navigate the complex but compelling subject area.

The foundation of fiscal policy and tax design for extractives is the grasp of resource economics. So I will begin with a quick survey of the principle concepts involved. The three I'll highlight are resource endowments, resource projects and resource rents. So to begin, let's consider the resource endowment, be it of oil, of gas or of minerals. In a physical sense, we know that there's a fixed amount of any resource that we consider of value today or even in the future. Unfortunately, there is really very little else that is certain about the resource endowment. Nevertheless, certain concepts have been developed to organize the way we manage that uncertainty and measure the size and value of resource endowments. So how much oil, copper, coal or diamonds do we have today? Even a purely physical assessment depends on making some assumptions about technology and that in turn depends on what is economically feasible. It may be surprising to learn that oil generally occurs in reservoirs under common pressure systems which when tapped by wells very rarely yield any more than 20% to 30% of the oil that's actually in place. Although enhanced recovery technologies exist, they are simply not able to stimulate that much additional production at an acceptable cost. And for many metallic minerals, the same is true. Although technology now allows recovery of metals from rocks in concentrations of less than 1%, going very much beyond this is just not feasible. Resource classification systems take these points into account, but for the purposes of making economic decisions on where and how much to invest, a further qualification needs to be made and this relies on resource prices. For any given resource amount that is technically available, there is likely to be a rather smaller amount that will have been fully measured and is available to be extracted economically at today's prices. This smaller amount is called reserves. In fact, even reserves are subdivided by levels of confidence into proven, probable and possible reserves. So if you think about it, exploration is the process of resolving resource uncertainty to the point of establishing proven reserves. It is risky, it's expensive and it takes years of effort. Now the main reason for elaborating this point is to underline that fiscal policy for extractives must contend with great uncertainty about anything other than the known stock of resources that has already been located and measured sufficiently to form the basis for economic extraction. So there you are, that's my take on resource endowments and some of the methods we adopt to address uncertainty and measurement.

Now let me move on to resource projects. A project moves through exploration to production and eventually to closure. This can also be thought of as a process of resolving uncertainty by applying technologies and taking risks. Hopefully the value of the resource that is eventually extracted will compensate the investor for all of the risks taken. Risk is highest when still trying to find the resources. It reduces once a discovery is made, but nonetheless remains significant while the resource is being measured and the feasibility of extraction evaluated. When finally, extraction begins, the initial investment may begin to be recovered. If all the assumptions that were made preproduction as to prices, technology and cost including taxes were accurate, profitability should be assured. That of course is not always the case.

Now let me move on to the third concept that I want to highlight, that of resource rent. This is a central concept in any discussion of resource taxation. Resource rents are the surplus value remaining over the resource project lifecycle after all costs attributable to generating that value have been deducted and that includes returns to capital. A characteristic of any stock of resource deposits that can be exploited profitably is that there is a very wide distribution between low, average and high cost deposits. Now as depicted in assumed chart, which shows the case of world copper supply, the resource price 'p' corresponds to the marginal cost of the marginal supplier. The marginal supplier is the supplier that is just able to cover its costs at prevailing prices. The lower of the two supply curves in the assumed chart will show short term running costs, whereas the higher supply curve also includes costs to remain in business and make adequate returns over the project life. What you will see from this is that low cost producers should be able to generate significant resource rents with the magnitude of those rents a function of the steepness of the supply curve. However, prices are not static; in this case the evolution of copper prices over a ten-year period, you can see the striking fluctuations in prices. So imagine a resource project lifecycle of say fifteen to thirty years. During this period there will be moments when of higher or lower profit margins or even outright losses due to these fluctuating prices. So I think it's important to keep in mind that, although we can represent the generation of resource rents in a static sense as depicted in the first assumed chart, the reality is that, over a project lifecycle, there is forever change and at the end of the day, the total amount of resource rents can only be known on the very final day of production when the resource has been fully exploited. As we will see later, fiscal regime design is guided by the idea that the state should be able to capture resource rents without distorting investment decisions since, these rents are a surplus. We will also see that in practice this is very hard to do. So far, I've talked about extractives as a whole, the question is do the concepts of resource endowments, resource projects and resource rents that I've addressed apply equally to oil, gas and minerals?

Fundamentally the answer is 'yes', although differences exist and these differences are very important to be aware of. The engineering parameters that define endowments of hard rock minerals and liquid or gaseous hydrocarbons are quite different and methods of resolving uncertainty through exploration vary accordingly. This influences the way in which risks are managed over the project lifecycle, but it is also important to realize that differences between, for example, oil and gas are just as significant. And differences between iron ore and say gold are not insignificant. Now, one difference that does generate quite a lot of interest is the apparent ability of states to capture more rent from oil than from most mineral projects. The underlying factor driving this observed factor is probably the steepness of the supply curve in the oil industry in which the lower cost producers in conjunction with exercising some degree of price control through OPEC, generate resource rents on the scale rarely found in the mining sector.

So, we covered some of the underlying concepts that are necessary to grasp the nature of fiscal policy-making and some of the challenges. What we covered I hope reasonably clearly, are resource endowments, resource projects and resource rents. This should provide the platform for moving on to address other issues.

FISCAL POLICY PRINCIPLES AND PRACTICE

Thought-provoking questions:

1.    How do a country's objectives shape extractive sector fiscal policy?

2.    How can fiscal policy affect the rate of resource depletion?

I want to discuss fiscal policy principles and practices, focusing particularly on the factors that tend to influence the content of fiscal policies for extractives. I'll highlight the economic context in which policies are made, some of the fiscal and non-fiscal objectives that are likely to drive policy, and the competition among countries that can influence policy and then I will end with an observation about depletion policy. I'll begin with the varying economic contexts found in different countries, which will influence extractives policies. First, I want to talk about the resource horizon. Oil, gas and mineral resources are non-renewable and will eventually be exhausted. That much is certain. But over what time horizon and at what pace, well that's completely uncertain. It's possible to project the depletion path of the known stock of resources of a particular quality under certain market and technology conditions, at least over a range of foreseeable scenarios. However, the reality is that there is no fixed stock of resources. But rather, as the known stock is depleted, it can be replenished through discovery or by new technology or simply by higher, real prices. What I mean by higher real prices, take the situation of the oil industry at the time when OPEC was formed; at that point, there was a step change in the relationship between the oil prices and the underlying supply costs, so much so that it stimulated the search for new oil in the North Sea and elsewhere that had hitherto been unviable. So through discovery, through the application of new technology or through the influence of changing real prices, we may find that the known stock of resources, can be replenished to offset the depletion. Of course, all three of these factors, discovery, technology and prices are inherently uncertain. And let's look at some practical implications of this. The resource horizon of Saudi Arabia, for example, is one of the longest with many of the first discovered oil fields back in the 1920s or '30s yet to be depleted and new discoveries and technologies are still expanding that resource base. Equatorial Guinea and Yemen, on the other hand, each foresee the exhaustion of their oil resources within the next two decades. And yet, two decades ago, they were only just beginning to exploit those resources. And then, there are those countries in which the oil resource base is only just beginning to be revealed. Ghana, for example, started that process just some ten years ago and in Sierra Leone, it's happening right now, so countries are in vastly different positions and this quite properly affects their fiscal policies. Good fiscal policy is founded on developing the best possible information on the resource base and the time horizon based on resource evaluations and the technical data obtained from companies that carry out exploration. However, a salutary lesson comes from the United Kingdom. Since the 1980s, knowledgeable people have been predicting that North Sea oil would run out and if any of them had actually been correct, the U.K. would long ago have stopped producing oil. Now, let me move on to the question of dependence on resource revenues. Again, countries have vastly different degrees of reliance on resource revenues, as well as very different development needs, access to other sources of development finance and differences in their absorptive and institutional capacity. For those countries that are or foresee being highly dependent on resource revenues, say 90% or more of public revenues, fiscal policy priorities will clearly be different than in countries with much more diversified revenue base. About 60 developing countries currently have relatively high levels of resource revenue dependency, of which about twenty, mainly oil producing, are extremely dependent. Then, another factor influencing fiscal policies for extractives are shorter term considerations of external and fiscal balances. These deficits in either or both of these balances can drive preferences for near term exports and revenue mobilization in preference to longer term growth of the revenue base and economic diversification. A few countries enjoy the rather rare luxury of receiving more on interest from resource funds accumulated over years than from resource exploitation itself. This interestingly is the rare position that Norway finds itself in, where it's actually making more in the interest payments from its sovereign wealth fund than from taxes paid by oil companies. Having discussed the economic context, I'll spend a little bit of time on some of the fiscal policy objectives as well as non-fiscal policy objectives which could in principle be set by governments. Of course, fiscal policies for extractives should reflect overall development priorities and management of the extractive sector, though in reality as we all know, policy rarely starts with a clean slate. It has to accommodate legacies that may sit uncomfortably with the ideal policies. I'll start with the obvious objective of revenue mobilization. It goes without saying that this a core objective of fiscal policies for extractives or the more so if there are very few viable alternatives available to the country to raise public funds. Now a rational economic approach would be to maximize the present value of net government receipts from the resource endowment. That, of course, leads to the question of what discount rate to apply to future revenue streams. The higher the discount rate, the less value placed upon future revenue streams. A high discount rate may also influence the choice of fiscal instruments, for example reliance on the taxation of profits alone may not satisfy the government's objective of at least receiving some revenue when production starts, hence, the typical resort to royalties on production, and sometimes and perhaps even better from a government point of view, the imposition of cash payments up front in the form of bonus payments to obtain resource rights which then can be cashed in from day one. By comparison, a government that wishes to encourage the growth of extractives and is not under immediate pressure to raise revenue may go so far as to provide incentives for investing in exploration. This can be done by allowing accelerated depreciation for example, or by allowing a portion of the abortive exploration costs to be written off against an income generating project. These all represent choices by the government to defer or reduce revenue receipts today in the expectation of greater revenue receipts in the future from a bigger tax base. Beyond the core objective of revenue mobilization, there may be some other factors that will influence policy, for example, the generation of economic spillovers and meeting of social goals from the extractive industries. So, should a government rely on revenue mobilization alone, or is there a role in promoting these economic spillovers? The government can selectively encourage such spillovers by allowing taxpayers for example to write off costs associated with local hiring and procurement of goods and services, or the building of infrastructure that also has some public benefit, or making corporate social investments for the benefit of the local community. The resulting revenues foregone, often termed tax expenditures, may be justified if there are commensurate benefits. Clearly, such a policy choice will be reflected in the fiscal regime design. It's worth considering that unless the government has a good track record of using revenue receipts wisely, there may also be pressure from civil society to place less emphasis on revenue mobilization and more emphasis on the wider benefits for citizens. This could for example include requiring domestic value addition or insisting on local hiring and procurement and providing tax breaks to encourage this. Of course another option where public accountability of resource revenues is weak is to demand that resource revenues are distributed to citizens directly in cash as in Alaska. Finally, in this survey of factors that may influence policy for fiscal regimes comes the question of ownership or taxation. A government can choose to achieve its wider fiscal and development goals through state participation in resource projects rather than relying entirely on taxation. State equity ownership is actually a more direct form of rent capture than taxation, but it is risk prone. It carries an opportunity cost for the use of scarce public funds. State equity risks can be mitigated in joint ventures with the private sector, but if there is a free or a concessional element in the shareholding, this operates as a de facto tax on the joint venture. So equity interests, especially outright ownership, tends to be prevalent where there is a known stock of large low cost resources, therefore offering a favorable risk reward for the use of public funds, such as in Saudi Arabia's oil fields. Even in such cases, reinvestment rates by state owned companies have tended to be low, resulting in diminishing returns on public funds and sometimes outright losses as in the case of Zambia when its copper mines were under public ownership. Moreover, in nearly all cases, the fiscal objectives of state participation can be met through good tax design and administration. But we have to bear in mind that state participation may be motivated not just by fiscal, but also by non-fiscal objectives. So I've covered a number of influences, underlying factors that may help to explain why the fiscal policies adopted by different resource rich countries do vary a great deal. Now fiscal policies for extractives are not of course set in a vacuum. In particular, there tends to be intense competition among countries to attract investment as well as fierce competition among countries to obtain rights to the best deposits. So the competitive environment itself is constantly changing owing to commodity cycles and the emergence of new players in the sector. While it is prudent to calibrate policy in accordance with these changes, constant shifting of policy objectives and the terms offered to investors will simply breed uncertainty and undermine the credibility of government policy. As in most things related to fiscal regimes for extractives, a fine balance has to be struck. Benchmarking of the fiscal policies and regimes offered against others is an important thing to do. For example, comparable countries ought to be those with similar geological potential, cost and operating environments, track record, regulatory capacity and actual or perceived country risk. Countries can therefore check at frequent intervals to see whether their terms are moving together with or against the prevailing trends among those countries that belong to the same peer group. I'll end this chapter on a particular point around the idea that fiscal policies are really in fact a form of depletion policy. A government has an opportunity to influence the pace of depletion and replenishment through the fiscal policies it adopts. A fiscal burden that reduces the portion of the resource endowment that is actually profitable to exploit will slow the rate of depletion of that endowment. This may be justified on a number of grounds, economic and non-economic such as environmental or social grounds. The government can also accelerate the exploitation of the resource endowment if it's willing to absorb some of the risks that otherwise would have been absorbed by the investor. Norway, for example, for some time was willing to forego taxes equivalent to 70 cents in the dollar of exploration by allowing write-offs for abortive exploration drilling. That was founded on a belief that the resource endowment remaining to be discovered was substantial and that by adopting such incentives that those resources would be discovered, measured and exploited and that Norway would be able to receive taxes into the future. Well, that completes my survey of a number of the principle influences on fiscal policies for extractives.

FISCAL DESIGN CONCEPTS FOR EXTRACTIVES

Thought-provoking questions:

1.    How does the progressivity or regressivity of a fiscal regime impact the government? How do they impact the investor?

2.    How does the large up-front capital expenditure required for most extractive projects affect fiscal regime design?

To design the fiscal regime that meets the objectives set by fiscal policies of extractives, it is necessary to understand the impacts of different fiscal instruments on the overall economic position of the investor. I'll discuss this and commonly used criteria for fiscal regime design. Private companies invest with the expectation of making profits commensurate with the risks associated with their investments and their cost of capital. In this respect, there's nothing exceptional about investors in extractives projects. The concepts that are important in order to understand investor decisions are those of the net present value of a stream of positive and negative cash flows over the project lifecycle and the internal rate of return of a project which will normally have to be met to justify investment. The features that distinguish investments in the extractives projects from others in other sectors are the long gestation period for projects. When the entire period from discovery to first production is counted, this may be several years if not decades and secondly, the heavy capital expenditures that normally precedes first production. These factors mean that the calculations of net cash flow are very sensitive to the discount rate used by the investor. With early outlays and discounted future revenues weighing heavily on decisions whether to proceed or not with the project. It follows that project appraisals are particularly sensitive features of the fiscal regimes that delay the timing of positive cash flows and extend the period during which capital remains at risk. The other important consideration is the premium applied by the investor to the chosen discount rate to account for risks affecting the cash flow. Investors will rely on the fiscal regime being clear, predictable and stable over the course of the project life, at least during the period when capital is being recovered. If the government's commitment lacks credibility, a risk premium is likely to be added, setting a higher bar for projects to get approved. Now, I want to spend a few moments on some of the main design criteria employed in designing fiscal instruments. First is the concept of tax neutrality. Namely, the principle that tax should not distort production decisions. A project that is profitable on a pretax basis should remain so on an after tax basis. Later, I'll illustrate how this criterion may be applied in the cases of royalty and income tax.

Second is the concept that the fiscal regime may be regressive or progressive. Put simply, a regressive fiscal regime is one that captures a higher share of the resource rent of a low rent project than the higher rent project and the progressive fiscal regime is the inverse of this. Progressive taxation as might be applied to say personal income, relies on the principle of taxing in according with the ability to pay, with the rich bearing hire taxes. While the merits of the latter principle may seem self-evident, it is striking that in very few cases our fiscal regimes for extractives are progressive. Many are in fact quite regressive. The result is that marginally economic projects bear a heavy fiscal burden and may be either deterred altogether or curtailed prematurely. By comparison, those projects that generate windfalls may be allowed to retain the lion's share of windfalls to the detriment of the government. Practical application of the relatively straightforward principle of progressivity is, however, difficult. First, much fiscal design tends to focus on individual fiscal instruments, not on the regime as a whole. When a progressive fiscal instrument, such as a profits tax with an escalating tax rate, is combined with one or another more regressive instrument, the overall impact may not be progressive. Second, progressivity is best understood on a project lifecycle basis. However, most tax instruments are designed to apply on a tax year basis, so even an escalating tax rate may not necessarily result in progressivity over the entire project lifecycle. Finally, the other criterion I want to mention is the common sense principle of simplicity. The principle that the fiscal regime should be no more complex than necessary to achieve its objectives, since a simple regime is more likely to be administered consistently and effectively and will impose fewer compliance costs on the taxpayer. This point is often lost on tax economists devising optimal fiscal regimes and negotiators seeking to close a deal. Complexity may be the inevitable outcome of governments that seek to meet multiple goals or focus simply on single instruments while ignoring the fiscal regime as a whole. That said, it is important to strive for simplicity with greater standardization in legislation and in model contracts. Finally, in order to design a fiscal regime according to the preceding criteria, it is necessary to have an economic model that will demonstrate the impact of the fiscal regime on a resource project over its lifetime. Such a model has to contain all elements of the fiscal regime. A piecemeal evaluation of individual fiscal instruments has limited value compared with an evaluation of the fiscal regime package as a whole. Such models can be used to simulate the effect of varying the composition of the fiscal regime and to evaluate how the fiscal regime responds under different assumptions about output, prices, costs and other project parameters.

FISCAL REGIME TYPES

Thought-provoking questions:

1.    What are the different types of fiscal regimes?

2.    Which revenue streams do they generate?

There is a great proliferation of fiscal regimes for extractives around the world, each trying to achieve the fiscal policy objectives we examined earlier. Even in a single country, there may be more than one fiscal regime offered. In the next few minutes, I want to discuss the principle types of fiscal regime and their main features.

Broadly speaking, there are two main families of fiscal regimes used globally.

The first is the royalty tax system. Named as such, because it combines a royalty on production with the taxation of profits and levies on inputs much as any other tax system would be designed. This system is prevalent in the mining industry, but it's also widely used in the oil industries, though much less so in oil rich developing countries.

The other fiscal regime family could be broadly viewed as a fee for service system, in which the state retains ownership, but rewards a contractor for its investment on the cost plus basis. Its best known form is the petroleum production sharing contract, under which shares of production are allocated first to allow the investor to recover its costs and then the portion of any residual production is allocated to the investor as its profit. There are very few instances when such regimes have been used in the mining industry, so this is essentially an approach found in the oil industry. Although these two quite different families of fiscal regime exist, in reality, the alternative types can be designed in such a way that the economic outcomes are virtually indistinguishable, even though the operational structures differ. So, I'll discuss in more detail, each in turn. There are usually five types of revenue stream the government can expect to receive from a royalty tax regime.

These are number one, levies on inputs into the project, such as customs, duties on imports, sales tax or VAT on inputs as well as taxes on wages and various other charges on inputs.

Two, a royalty that's charged on production either as a fee per unit of product or on the basis of the sales value. Third, a percentage of net income of the project after deducting operating costs, depreciating the capital, deducting royalties and other input levies either using a conventional income tax or resource specific profits tax. Fourth, withholding charges on payments of dividends and sometimes interest to shareholders, and finally miscellaneous minor fees such as area rents, license fees, environmental fees, property taxes and the like.  You will see the timing and approximate magnitude of these different revenue streams in relation to the overall project lifecycle. Input taxes are the earliest to be paid, followed by royalties. And, income taxes come later and are typically the largest revenue streams so long as the project is making a normal level of profits. Some countries have additional taxes on profits to increase tax receipts when windfall profits are being made. If a project is unprofitable, then there is at least an assurance that some royalty will be paid. The royalty is commonplace in such fiscal regimes, however, as this chart shows, it is both regressive and non-neutral. Its burden falls disproportionately on low rent projects and reduces marginal supply. Moreover, income tax relief for royalty which is normally available is really only of value to projects that have taxable income, thereby aggravating the regressive effect on marginal production. It has become generally accepted that so long as royalties are imposed at modest rates, they are unlikely to be too distortive. Moreover, royalties traditionally being regarded as a minimum payment due to the owner of the resource for the right to deplete that resource. If there were no royalty payable, the state could face a situation in which it receives no payment from a loss making project, even though that resource has been fully depleted. Typically, royalty rates are in the range between say 3% to 6% of the gross value of minerals and between 5% and 12% of the gross value of oil, though there are outliers of course in both sectors. Nevertheless, when resource prices drop suddenly, pressure usually mounts on the government to differ or even cancel royalty in a bid to prevent job losses and project closures. It follows that levies on profits are preferred by investors to royalties, although most will accept a modest level of royalty. Depending on the effective rate of profits tax after allowing for depreciation, such taxes are much closer to being neutral. A normal flat rate income tax in fact will impose a proportionate burden on both low and high rent projects. Now the objective of optimizing tax revenues by targeting resource rents calls for the tax burden to be low on the lower rent projects and higher on higher rent projects. As we discussed earlier, such a progressive regime is, from an economic point of view, desirable in which about one-third of the rent is captured from the lowest rent project, whereas two-thirds is captured from the highest rent project. The type of tax most likely to achieve this capture of rents is a resource rent tax, which will be discussed further.

Now, I'll move on to production sharing. This approach, developed in late 1960s in the oil industry, it helped then to address a government goal,  that was quite prevalent, of retaining ownership of resources and obtaining a share of the global crude oil market, which in those days tended to be dominated by a few western countries. The critical point about production sharing was that it also needed to offer incentives to private sector companies to continue to explore for oil. Under production sharing, oil is first allocated to the investor to recover its costs. This is called cost oil. And then second, to share whatever oil remains on the basis of an agreed split between company and the government. This is called profit oil. In this pure form production sharing in fact functions very much like profits tax. Although simple in principle, production sharing regimes have become quite varied. Cost oil, for example, may be subject to a ceiling, having an effect rather like a royalty. Profit oil allocations may be linked to a variety of sliding scales to apportion increasing shares to the government in line with oil field size for example, or profitability. In fact, one survey of production sharing contract terms undertaken by the IMF found thirty-five different ways to structure cost oil and profit oil. Moreover, in many countries, production sharing contracts are actually embedded within the prevailing royalty tax system of that country, thereby adding considerably to complexity. While production sharing may have been borne of the concern of host countries to assert control over resources and oil markets, investors in fact have found that the system offers greater simplicity often than royalty taxation and some assurance of stability. The latter arises because of its contractual nature and typical provisions for offsetting changes in the wider tax system with contractual adjustments of the profit oil splits. Having discussed the two broad families of fiscal regimes, namely royalty tax systems and then production sharing systems, which is the main manifestation of a fee for service system, I want to mention briefly state equity which we mentioned in an earlier chapter when looking at ownership and taxation. The state might opt to participate in a resource project on the joint venture basis. And this should not be ignored in considering the design of the fiscal regime. If the terms of such participation are free or concessional, that is to say not pari-passu with private investors, then the terms should be considered as part of the fiscal regime, since they impose a burden on the investor. For example, a state interest in profits with no associated liabilities is the equivalent of imposing an additional withholding tax on dividends. Well, I've covered the two main families of fiscal regimes, namely royalty tax and production sharing agreements and then made sure not to forget about state participation and its impacts on fiscal regime design.

FISCAL REGIME DESIGN CHALLENGES (I)

Thought-provoking questions:

1.    How can governments capture resource rents under a wide range of market conditions?

2.    What are some of the common challenges that arise in fiscal regime design?

Having described the main features of fiscal regimes found in the extractives industries, I want to discuss some of the recurrent design challenges facing policy-makers and their tax advisors. First of all, let me talk about what I call flexibility. As we've seen one of the main factors driving the diversity of fiscal regimes is the attempt to capture resource rents from a very diverse resource space under a very wide range of market conditions. If a fiscal regime is too rigid, either the government or the investor may be unsatisfied with the outcome and seek to redefine the terms. Thus, flexibility and fiscal regimes for the most part has been accepted and practiced. There is in fact a long tradition of designing fiscal regimes to anticipate likely levels of resource rent ex-ante. Governments often offer distinct fiscal regimes for example for different resource types or areas according to their prospectivity. For example, Botswana offers a fiscal regimes tailored to its world class diamond deposits that is quite different from the regime for other minerals. Indonesia's terms for oil vary depending on the basis of whether projects are located in prolific basins or in high risk frontier areas or in mature and depleting areas. Even individual tax instruments may be designed to reflect different economic conditions. Examples include royalties for off-shore oil areas where the royalty rate is set in relation to the water depth. This is done on the basis that as increasing water depth, costs are likely to be higher. And so the royalty rate is set, anticipating the impact that that would have on profitability. In production sharing, the profits splits may be linked to oil field size or the predicted level of costs. In all of these cases, there is an attempt to build flexibility into the design of the fiscal regime based on predictions of what economic factors will drive the level of rents generated. The alternative approach is to base fiscal regime design on ex-post outcomes. In other words, on the ability to pay. For example, rather than using a flat rate income tax, the rate could be on a sliding scale linked to dollar amounts of profit generated or the profit margin in the particular tax year. In the case of production sharing, some regimes link oil splits to what's called an R-FACTOR, which measures the ratio of cumulative income to cumulative costs since the project started. Rates of return taxes represent the closest to a pure resource rent tax with the tax triggered only once the defined project rate of return has been exceeded. That point is determined by compounding costs and revenues at an interest rate equal to the target rate of return. Once the net cash flow becomes positive, the target rate of return will have been achieved. Either a single or several rate of return thresholds can be set at which progressively higher tax rates are applied. This type of tax is used in some fifteen to twenty countries around the world for both oil and mining. An ex post approach to targeting resource rents is superior to an ex-ante approach on economic grounds. However, its effectiveness does depend on being able to administer a system of audit to verify that the outcomes of the project, namely all historic costs and revenues, have been declared correctly. Now, let me move on to tax incentives. This is another challenging area. There is a long tradition of employing tax incentives to lure investors in the heat of competition or because government officials are persuaded by investors that the project will not be viable without such an incentive. Historically, tax holidays, special depletion allowances and investment uplifts have been granted that either eliminate or significantly reduce the tax liability. The record of such incentives is poor and the public policy consensus has moved against them unless they are linked to explicit, enforceable performance obligations that offset the taxes that are foregone. Earlier on, I explained that the non-fiscal objectives connected with for example local hiring and procurement or infrastructure provision might nevertheless justify special incentives. One incentive that has wider acceptance and is found quite commonly in tax codes for extractives is accelerated depreciation. The policy premise for accelerated depreciation is the long gestation period of extractives projects and large upfront capital outlays, a large portion of which are in the form of intangibles like drilling costs. The ability to write off accumulated exploration and development costs at a rapid rate, which is what accelerated depreciation offers, responds to an investor's interest in limiting the investment recovery period. The concession made by government is to defer tax receipts, but not to lose them altogether. Ringfencing is a tax measure that has particular relevance to fiscal regime designed for extractives, so I want to spend a bit of time on this. This is a rule that limits what source of income and what expenses can be consolidated for purposes of determining taxable income. Given the revenue maximizing principle of rent capture in the extractive sector, it does follow that only the income and the expenses of a single project should be consolidated for tax purposes. Having said that, some governments do accept that the ringfence could be relaxed. This would be to enable non-project expenses to be deducted from project income as an incentive for existing resource producers to invest further in exploration and the development of new projects. This would have the effect of delaying tax receipts from current resource producers. Note, however, this particular benefit would only accrue to existing producers. It would not accrue to new entrants into the sector for which alternative forms of incentive might be required. Now in considering the challenges of tax incentives, governments do have to be mindful that any tax incentive, which results in the differential tax treatment of related activities, could generate strong incentives for transfer mispricing. For example, affiliated entities could be established under common ownership to take advantage of different tax regimes on mining and on mineral processing. If the processing entity were tax free, mine products would be transferred to it at cost, thereby transferring all profits to the tax free processing entity, a classic case of domestic transfer mispricing.

Another example is when generous interest rate deductions encourage corporate financings to be structured on the basis of shareholder loans to related project entities with inflated interest rates. Since servicing debt ranks ahead of tax liabilities, this type of arrangement can erode the tax base. It should not be forgotten that investment behavior is influenced by many other factors, both fiscal and non-fiscal. There is a temptation to overemphasize fiscal regime design in the belief that it alone will drive investment and rent capture outcomes. So I want to close this chapter by highlighting those other factors that underpin the performance of the fiscal regime. These include, the first, predictability. Can an investor or indeed a tax collector reliably model all the elements of the fiscal regime and how they will work? Second, stability. Even assuming the fiscal regime is predictable, will it remain the same and what provision, if any, has been made to adjust to any changes? And thirdly, compliance. How much effort will it require to comply with the fiscal regime and in the event of dispute, is there a reliable and fair method for resolving these? It is in fact arguable that an investor would prefer a high burden fiscal regime that meets those three conditions over a low burden fiscal regime that does not.

FISCAL REGIME DESIGN CHALLENGES (II)

Thought-provoking questions:

1.    How should revenue collection and administration influence the design of a fiscal regime?

To end this series of discussions on fiscal regimes for extractives, I'll discuss how such systems are actually administered and revenues collected and some of many associated challenges.

The first point to note is that the design of fiscal regimes tends to be dominated by tax policy decision-makers and their advisors and negotiators of contracts and their advisors but very rarely, revenue administrators themselves. The classic outcome of a negotiated fiscal regime is that of the tax commissioner receiving a copy of the clause which his office is supposed to implement without ever having been party to the negotiations and never seeing the entire agreement owing to its secrecy. Clearly, that won't do. Good practice would be to have tax administrators engaged in both the design and the implementation of fiscal regimes. It is worth considering whether administering a fiscal regime for extractives is any more challenging than administering those for other sectors. I think there are three types of challenge that are quite significant in the case of extractives. First of all, there are the special characteristics of the extractive sector itself. This includes multiple commodities and pricing arrangements, a hugely varied project costs, diverse participants through the exploration, development and the extraction cycle, and a prevalence of multinational corporations, globalized value chains and complex project financings with a high use of affiliated transactions. Second, there is the complexity of administrating nonstandard fiscal regimes comprising multiple instruments. These variations result from statutory adjustments, variety of tax incentives and project specific negotiated terms. Moreover, older fiscal terms may be grandfathered by a stabilization clause, leaving the administrator having to become an expert on a variety of different provisions, some of them may no longer be in wide use. And then third, revenue administration is typically quite fragmented with jurisdiction over direct, indirect and resource specific levies being held by different agencies. In the case of oil, production sharing is typically handled by the oil ministry if not the national oil company. And tax liabilities may be assumed by the national oil company on behalf of the oil company. In the important area of audits, auditors may rely heavily on data that is only available from the sector regulator or the national oil company, especially related to physical verification of the products produced, transported and sold and both tangible and intangible assets used in the production process. A combination of these three sets of challenges can result in a situation in which revenue agencies lack the know-how, access to information and the resources necessary to measure and value resources to classify and verify costs and to safeguard against transfer mis-pricing on a reliable basis. Now, is this problem only in the poorest countries? No. Take Indonesia, which has been a major producer of minerals for the past half century. World Bank work there showed that the government may be only receiving 50% of the royalties due to it on coal. The reasons for this great leakage are the ones that are common to many countries. It's a mix of lack of coordination among national agencies, and between national agencies and subnational agencies, as well as rent seeking and corruption. Data collection and sharing, control and compliance processes and audits both physical and financial have all been compromised as a result. A full discussion of how to strengthen revenue administration in light of these challenges is beyond the scope of this training module, however, measures that do offer some promise include the following.

First, moving from tax type segmentation to taxpayer segmentation and setting up large taxpayer offices.

Second, introducing risk based audits to better deploy scarce resources and target significant sources of leakage.

Third, unbundling the functions of the sector regulator, especially promotion and licensing from revenue assessments and collection which might otherwise be conflicted. Fourth, focusing on the strengthening of how international tax issues are handled such as transfer mispricing and treaty shopping. And finally, limiting, as far as possible, the extent to which the assessments made by tax administrators and audit rely on data that is hard to come by and verify, including the use, where appropriate, of comparable data, reliable indices of prices, costs and other relevant project parameters that will relieve the auditors and revenue collectors from embarking on very costly and time-consuming inquiries and reviews. So the final word on revenue administration relates to transparency and its role in promoting greater accountability and performance of revenue agencies. Demand for disclosure of revenue flows from companies to revenue agencies as required under the EITI standard and new home country reporting requirements is increasingly placing the spotlight on government institutions not only to account for revenue receipts and their use, but also to demonstrate the efficient performance of their revenue assessment and collection functions. And now that details of fiscal regimes are more open to public scrutiny and the details of contracts are coming into public view, fiscal regime design is also under increased scrutiny. In fact, there is growing interest in the open source modeling of fiscal regimes that can be deployed by a range of stakeholders, not just companies and governments, to judge whether the fiscal regime for extractives is actually meeting the stated policy goals of the government. That brings to the end this discussion on the extractives fiscal regimes.

Thanks very much.

Prepared by Kiiza David Maclaren

​MBA PNG (Amity), B.STAT (Hons-Muk),DIP CIVIL (UNEB), IELP (North Carolina)

GEM (DTU-Denmark),NRSD (SDG Academy)