As noted, the stakes and risks in extractives projects are high. What guides investors in choosing projects are two factors: first, the underlying profitability of the project, both at the cash flow stage, and after taxation and the fiscal regime have been taken into account; second, how revenue flows stack up over time – on the principle that time is money.
The Investor Discount Rate
The way time is factored into financial analysis of a project is through the discount rate. Investors make projections of how much profit they will make in the future, and then analyse the times during the life of the project at which they expect these flows. Then they apply a discount rate to say that the further out in the future profits are, the more they should be discounted by comparison with the investment that needs to precede it to make it happen. The discount rate is expressed annually. For example, if a discount rate was 10%, an investor would calculate that for an investment of a million dollars, they would need to make a profit of over $1.1 billion by the end of a year before they started to generate returns that met their minimum return on capital. If it took two years for revenues to arrive, the minimum necessary return would be $1.21 billion – the same 10% applied in each of two successive years, with the added wrinkle that it is compounded, so that in the second year, the 10% would be applied not to the original billion dollars but to $1.1 billion, since the discount rate implies that is the total that should have been generated by then.
One way of looking at a discount rate is as a self-applied interest rate, applied to an investment that one is directly involved in. A normal investor might look at investment options, try and calculate future profits, and then subtract, or discount, general interest rates in the banking system. The thinking behind this is that any sum of money could earn such an interest rate safely in a bank, and that other investments therefore need to give higher returns. The “real” return of these other projects is therefore the extent to which they outperform the general interest rate.
That general interest rate would also form a part of a discount rate, since the investor would either be investing their own capital, which they could equally well deposit in a bank, or borrowing it, in which they case they have to pay interest on the loan. But it will also include all other factors which add up to what the opportunity cost is of investing in this project as opposed to another one. These could include: what are other mining or oil projects available to an investor – large multinational groups run internal competition processes between new projects bidding for management attention and shareholder capital; there could be expansion options within existing projects, or even investment not in increased production but in processes and technologies that would cut costs, and thereby increase profit margins on the same turnover. Conversely, what are the risks of one project relative to another? The same gold mine in terms of project economics might look very different in terms of “above ground” risk depending on whether it was located in a politically stable and relatively prosperous country or, by a poor and politically turbulent country. In other words what is the “country risk” of a given project?
All of these factors can be packaged into an indicator known as the Weighted Average Cost of Capital (WACC). Technically speaking, this analyses what an investment is costing in terms both of debt taken on to finance it, and the cost of any core capital a company is investing. In theory those calculations in turn should have factored in things like country risk, since they should affect both the cost of borrowing and the cost of capital.
WACCs are compiled across industries and the chart below, compiled by Aswath Damodaran, a financial analyst and professor at New York University, shows how different sectors compare.
In reality there are many other wrinkles, such as whether these rates are compiled in terms of “real” interest rates (stripping out inflation) or “nominal” interest rates, and whether data are equally available from all parts of an industry, particularly oil and mining where so many projects take place in remote parts of the world. The WACC rates here are indicative of the way company discount rates are formed, and should not be taken to be definitive, and companies often keep the discount rates they are using in particular projects, and the way they were compiled, a closely guarded secret. Ultimately, discount rates, like rent, are a concept which it is relatively easy to conceptualise but often extremely complex to formulate and execute.
One consequence of factoring in not only much profit, but when that profit arrives in the life cycle of a project, is that analysis may yield results that are counterintuitive. For instance, one project may have lower overall returns in cash terms than another, but be a more attractive investment, either because the profits are calculated to arrive sooner, or because, for one reason or another the company is applying a lower discount rate.
Another key factor to bear in mind here is the longevity of extractives projects. Since a discount rate, like an interest rate, is expressed annually, and compounded, the effect of short-term costs can be dramatic in projects which measure decades. The Libra offshore project referenced above requires $90 billion of investment over the next few years. The size of the oil field is considered to be enormous, and the project could well last into the middle or even the latter half of the 21st century. But how much profit would the project need to be earning in the 2040s, or 2050s, to justify tens of billions of dollars of investment now? It is another area in which impact is non-linear. It is therefore no surprise that companies are extremely sensitive to costs incurred up front, and that fiscal tools such as signature bonuses, which occur right at the start of a project, are unpopular.
Indicators of Profitability
The two core indicators used by investors to measure profitability, the Net Present Value (NPV) and the Internal Rate of Return (IRR) both relate to the discount rate. We examine each of them in turn.
Net Present Value
The Net Present Value is exactly what it says: a calculation about an investment, or series of investment in the future, which yield revenue streams, also in the future, expressed as a value of money right now today.
Perhaps the easiest way to think about it is as working like interest rates do, but in reverse. Say an investor puts $100 into a bank account where it earns 10% interest a year. After one year, the investor would have $110. After two years, the investor would have $121. Notice $121 not $120. This is because the interest compounds – interest earned one year gets added into the base on which interest is calculated the next. So in the first year, 10% of $100 is $10. But that interest from the first year is added into the base for the second year, the 10% interest rate is applied on $110 – not the original $100. This makes interest of $11, not $10, in the second year and so we arrive at $121 returns after two years.
Given these facts, we can say that $100 today will be worth $121 in two years time if invested in this way.
All the NPV does is to reverse the direction of this calculation. Instead of using an interest rate to calculate the future value of a current investment, the investor uses a discount rate to calculate the present value of future investments. If in this case the discount rate was 10%, in other words the investor needed a 10% return on capital each year, then the same conditions which lead to a value of $121 in two years time lead to a Net Present Value of exactly 0 (once the original $100 investment has also been subtracted). That $121 withdrawn two years from now, in other words, is the same value as $100 now – because time is money.
Obviously everything here does then depend on the discount rate used. With the same initial investment and subsequent returns, the NPV turns positive with a discount rate of anything less than 10%. For example, the same return of $121 in two years would yield an NPV of $9.75 at a discount rate of 5%. If, on the other hand, the discount rate was set higher, at 12%, (because the investment was perceived to be risky), then the NPV would be -$3.75. And, it is important to note, that negative NPV has been achieved with the same results in nominal terms: the investor is still estimating investing $100 now and receiving $121 in two years time.
No investor will invest against a negative NPV, as it implies they could earn more doing something else with the same amount of money.
Another key feature of the NPV is that it provides apples-for-apples across geography, time, and commodity. A company could calculate NPV against an iron ore prospect in West Africa to last 35 years, and a gold mine in Indonesia lasting only eight years and the results would be directly comparable.
From a government’s point of view, it is also very important to focus the time scope of an NPV calculation, and to distinguish between life cycle on the one hand, and point forward project economics on the other. This is because price volatility and the long lifetimes of extractives projects could combine to give radically different NPV results from the same project. When commodity prices crashed in 2014, there were cases of companies presenting negative NPVs to governments to persuade them to relax terms of royalties, or taxation. But if such NPVs were determined over the life cycle of a project, including capex that had already been sunk, then they would not be the relevant metric. Instead, the government should be looking at the point forward NPV – is this mine, or oil field, going to yield an operating profit from this point forward? The life cycle metrics were estimated by the company before it took its final investment decision. If that decision has been taken, and the investment is sunk, it is only the point forward economics, expressed in the NPV, which will determine whether continued production is viable or not.
The last point about the NPV to note is that, precisely because it tracks flows of money against time, the same absolute cash flows over a whole project lifecycle could yield different NPV values depending on the life stage of the project. Suppose, for example, the $121 was returned to the investor over the course of two years but in a different time sequence. Instead of receiving nothing for 24 months and then $121 at the end, the investor got $21 after 12 months, and then the return of the original $100 investment at 24 months. Although the figures are the same in absolute terms, the NPV is now positive (slightly, at $1.58) because the investor got more payback sooner.
Internal Rate of Return
The IRR is the twin sister of the Net Present Value. Instead of being an absolute sum of money, it is expressed as a percentage in the same way an interest rate would be. So in the example above, the IRR is 10%, just as the interest rate is. It is the fact that the IRR is the same as the discount rate which makes the NPV 0 in this case. If the IRR is higher than the discount rate, the NPV will always be positive, and if it is less than the discount rate the NPV will be negative. So the IRR and NPV are really different expressions of the same calculations.
As a rule of thumb, investors are often looking for an IRR of 15% or more, after all costs including taxes. Though there is no standardised rate.
Pre- and post-fiscal profitability
Investors will analyse profitability both before and after anticipated tax liabilities. In cases where terms for taxation can effectively be negotiated (such as an individually negotiated contract, for example), the comparison might lead them to specific suggestions on royalty rates, cost recovery and depreciation procedures, or a host of other mechanisms that combine to form the “fiscal regime”. Without diving too much into detail, which are dealt with in the separate paper on Petroleum and Mining fiscal regimes, it is important to understand that these different terms interact with each other in a way that makes it impossible to consider each in isolation. For example, the higher the royalty exercises on gross sales, the lower will be a company’s profits, and therefore any revenue streams to government which are based on profits such as corporate income tax, a special sector tax, or resource rent tax. This is why it is impossible to truly understand the economics of any given project without a detailed financial model.
Governments of course need to run their own analyses of project economics like companies do.
Conventional financial analysis often assumes the same discount rate to be applied to a project and then run against revenue flows to both the investor and the government alike, with only minor divergences For instance, since in many projects the government makes no direct financial investment, an Internal Rate of Return cannot be calculated.
One key question is: are or should the considerations be exactly the same for a government as they are for a company? Since a discount rate is supposed to integrate fully opportunity cost, using the same discount rate as a company would imply the government faces the same range of risks and opportunities. If this is not the case, should a government construct its own discount rate? And if so, how?
The most broadly used metric to calculate returns to government is what is called “the government take”. This takes the overall profits generated over the lifetime of a project and calculates what percentage of them go to the government, and what percentage to the investor. The International Monetary Fund deploys a specific implementation of government take in its modeling for governments known as the Average Effective Tax Rate (AETR).
Government take provides comparable numbers between one project and another. A bigger question is, are the projects themselves comparable? For instance, can we predict that because Project A has a higher government take than Project B, it is a “better” project for a government, and was “well negotiated”?
And the answer is: not necessarily. As one simple example, government takes tend to be considerably higher, generically, within the petroleum sector than they are in mining, in a way that even IMF economists have stated they cannot fully account for. Then there is the fact that price and cost volatility have constant, unpredictable and non-linear impacts on government take, with the further complication that two otherwise identical projects could be affected differently in their government take depending on what the terms in the fiscal regime are. Third, since profitability and risk can vary so much within the same industry, it is natural that projects could have different government takes. In the Middle East, where large oil reservoirs are proved and production costs remain low, the level of risk for investors is much lower, leading to governments being able to impose mechanisms that capture a higher proportion of the economic rent, and create a higher government take. In Iraq, for example, the government’s take in the massive oil projects it has signed with multinationals since 2009 average over 90% - higher than in most other countries. But is this because the Baghdad government has negotiated and run these contracts better? Or is it because they started from a much stronger negotiating position, because of the scale of their reserves and low production costs?
The Impact of Project Financing
One factor which has been underestimated by governments has been the impact of project financing and incorporation structures on revenue flows, and in particular tax liabilities. Because, as noted above, even well leveraged companies will borrow financing rather than invest core capital when they can, and because financing costs are often tax deductible, and many other tax revenues depend on which corporate entity where is engaged, governments can often get a nasty shock as projects develop. Time and again in the last few years, officials have found that a project which looks from the outside as though it should now have reached profitability, and should therefore yield corporate income tax or some other profit-related taxation is not doing so because of these structures. Positive cash flows do not translate into accounting profits because the operating entity itself may be loss-making, because of complex project finance and transfer pricing arrangements with other entities. So the project as a whole may be profitable at one and the same time as the taxable entities within it are not.
Integral then to a government understanding what its actual revenue flows look like from an extractives project is understanding the exact corporate structures combining to invest and operate it. In this sense, “beneficial ownership”, covered in a separate paper, must also count as necessary information to reach a fully informed evaluation.
Materiality of different factors
With so much uncertainty, and so many potential variables affected by so many factors, many of which may lie beyond the control of anyone directly involved in a project, managing complexity becomes a key consideration for many governments. In the context of project economics, then, it is important to understand that different factors of uncertainty can have a significantly different material impact on revenue flows.
Future price is the single most material factor, and of course impossible to predict. But impossible to predict does not mean impossible to manage – with financial modelling governments, like companies, can understand how a range of different price points will interact with profitability and taxes. In terms of cost categories, development costs tend to be very material, because they tend to be both large and “frontloaded” – preceding revenue generation, for example. Other categories like exploration costs might be less material, while operating costs become gradually more material to predictions in the life of a project to the extent that they start to represent a higher and higher proportion of total anticipated remaining costs.
Not all revenue streams are equal. Public attention, or a local EITI process, might focus significantly on land surface rentals, or commitments to training costs, even when these represent tiny proportions of projected revenues. Accountants can subject cost recovery structures to detailed analysis of capital depreciation rules even if they have marginal impact on overall government take.
Coming to an understanding of how to weight various factors project by project can make overall management of project economics easier for governments.