Introduction to Petroleum Economics 9781613994931

Citation preview

Introduction to Petroleum Economics

Introduction to Petroleum Economics

By Chris Hinkin BS mechanical engineering, University of Bath, UK MBA, Aston University, UK

Society of Petroleum Engineers

© Copyright 2017 Society of Petroleum Engineers All rights reserved. No portion of this book may be reproduced in any form or by any means, including electronic storage and retrieval systems, except by explicit, prior written permission of the publisher except for brief passages excerpted for review and critical purposes. Printed in the United States of America.

Disclaimer This book was prepared by members of the Society of Petroleum Engineers and their well-qualified colleagues from material published in the recognized technical literature and from their own individual experience and expertise. While the material presented is believed to be based on sound technical knowledge, neither the Society of Petroleum Engineers nor any of the authors or editors herein provide a warranty either expressed or implied in its application. Correspondingly, the discussion of materials, methods, or techniques that may be covered by patents implies no freedom to use such materials, methods, or techniques without permission through appropriate licensing. Nothing described within this book should be construed to lessen the need to apply sound engineering judgment nor to carefully apply accepted engineering practices in the design, implementation, or application of the techniques described herein.

ISBN 978-1-61399-493-1 First Printing 2017 Society of Petroleum Engineers 222 Palisades Creek Drive

Richardson, TX 75080-2040 USA http://www.spe.org/store [email protected] 1.972.952.9393

Acknowledgments My most grateful thanks go to Brendan Carey, Dr Fiona Macmillan, Abigail Martin, Norman Selley and Sharon Sievert for their reviews of, and improvements to, this book’s draft. Its final version would have been much the poorer without their help. And many thanks also to Jane Eden, Leah Guindon and Doreen Chin who expertly shepherded the manuscript through the SPE’s publication process.

Dedication Dedicated to my parents.

Foreword I recall with very fond memories my time as a student at Imperial College, London in the late 1970s, studying on the MSc degree course in petroleum engineering. Like all the other students, I absorbed everything I needed to know to begin an exciting career as a petroleum engineer—how to drill wells, analyze the data obtained from those wells (cores, logs, drill stem tests, etc.), understanding the basics of geophysics and seismic interpretation, the mapping of fields, learning the wonders of material balance, poring over the intricacies of offshore platform and pipeline designs, figuring out how the fledgling technology of reservoir simulation worked, etc., etc. Almost as an afterthought, every week we had a 1-hour lecture on petroleum economics, which we all grudgingly attended, regarding it as distraction from the much more interesting and important subjects we were learning about. Entering the industry immediately thereafter, it quickly became apparent that economics were a huge focus for petroleum engineers. I so wished I had paid more attention to those 1-hour lectures at Imperial College! Chris Hinkin, the author of this book, and I worked together for 10 years, between 2005 and 2015. He taught me the importance of deep expertise and understanding of petroleum economics. He created literally many hundreds of millions of dollars of value for the company we worked for by applying his profound knowledge and his painstaking attention to details. Chris demonstrated the need and criticality of petroleum economics experts in our business. I wish I’d come across him 25 years earlier in my career. I’m delighted Chris has decided to share his skills with a wide audience by writing this excellent book, which I wholeheartedly recommend to anyone wishing to improve their understanding of the subject—a must read. Howard Paver Senior Vice President Hess Corporation, New York City

Contents Preface Introduction Some Housekeeping Chapter 1: Introducing Cash Flow Chapter 2: What Is It About Cash Flow? Chapter 3: The Relevant Cash Flows Oil Companies “Know” Chapter 4: The Relevant Cash Flows Oil Companies Have to Calculate Chapter 5: Net Cash Flow Chapter 6: Introducing Discounting Interlude Chapter 7: Cash Flows in Practice Chapter 8: More on Discounting Chapter 9: Other Financial Measures Chapter 10: Inflation and Real and Nominal Cash Flows Chapter 11: Foreign Exchange Rates Chapter 12: Building and Using Discounted Cash Flow Models Chapter 13: Incremental Economics Chapter 14: Exploration Chapter 15: Sensitivity Analysis Chapter 16: Acquisitions and Divestments Chapter 17: Lease vs. Purchase Decisions Chapter 18: Presenting Your Work Chapter 19: Going About Your Business Chapter 20: Some Advice for Non-professional Petroleum Economists Chapter 21: A Final Word Appendices Index

Preface Every year, oil companies spend eye-watering amounts of money searching for and producing oil, and take breathtaking risks that they will be rewarded. And at the heart of deciding whether they would be wise or foolish to invest their money in drilling an exploratory well, say, or building an oil pipeline are petroleum economists whose job is to determine whether their company should grow richer or poorer by doing so. So, clearly, what petroleum economists do in oil companies is important. In fact, surely little else that goes on in an oil company is more important.

Introduction I wish someone else had written this book and I could have read it when I became a petroleum economist nearly twenty years ago. My background then was relevant enough that both my employer and I were optimistic that, with some training, I would succeed in my new role. So I read a few books and went on some courses to help. And help they did. But only up to a point and not completely. In fact, they reminded me of Mrs. Fotheringham. Mrs. Fotheringham was one of my math teachers at school, who taught by providing instructions: if the question asks this, you do this; if it asks that, you do that; which worked inasmuch as I passed tests. But I didn’t always understand why I was doing what I was doing. And the same was true of those books and courses. They provided me with instructions but not always an understanding. Since then I’ve read more books and gone on more courses that have helped me up to a point but not completely and have reminded me of Mrs. Fotheringham. But never mind, because after nearly twenty years as a petroleum economist (that has proved to be very far from the relatively small step becoming one was meant to be), I now understand. So what follows is the book that I wish I could have read twenty years ago. Chris Hinkin

Some Housekeeping Before we start, I want to make sure that anyone going beyond this point knows what’s in store for them. I want to be clear about what I mean by petroleum economics; I want to be clear about where I’m going to draw the line regarding what I believe is necessary to know and what isn’t; and I want to be clear about how I’m going to say what I’m going to say. That way, no one will be disappointed expecting this book to be something it isn’t. So let’s begin with the most important question: what is petroleum economics?

What Is Petroleum Economics? I gave up telling people that I’m a petroleum economist a long time ago after doing so had killed too many conversations stone dead. In a change of tactics since then, my conversations have enjoyed a much better survival rate by my telling people that I’m an economist in an oil company. Why the change worked, I don’t know. Clearly, in some way, “economist in an oil company” says something more meaningful than “petroleum economist” does. The trouble is that while conversations survive now (albeit, often, left wounded and suffering), they usually take off in a direction that isn’t at all relevant to what I do. People assume I forecast oil supply and demand, or I estimate countries’ economies’ resilience to oil price, or I’m an oil trader. But I do nothing of the sort. In fact I’ve already alluded to what a petroleum economist’s job is in the earlier Preface, which I’ll repeat here, paraphrased: A petroleum economist’s job is to determine whether his or her company should grow richer or poorer by investing its money in, for example, drilling an exploratory well or building an oil pipeline. I’ll talk about certain aspects of that sentence later on, like why “should,” why the sense of uncertainty? And how do we measure “richer or poorer’? But the definition is good enough for now. So petroleum economics makes the case to oil companies to spend money on something because to do so should make them better off; and, possibly even more important, it also makes the case to them not to spend money because to do so should make them worse off. Of course, cases are also made at the same time to do one or the other by other parts of the company, which may relate to political issues, or the company’s own strategy, or tactical maneuvering to thwart a competitor; and decision makers have to consider all these things too in coming to their decision. But this book is about petroleum economics, so I’m going to stick resolutely to that topic and not get distracted.

There’s just one more thing to be clear about here, which is where in the oil industry petroleum economics applies. The oil industry can really be divided into at least two subindustries: the upstream part and the downstream part. The upstream part relates to oil companies’ activities in exploring for, and producing oil from, oil fields, and the downstream part relates to their activities in refining what the oil fields produce and selling products such as petrol (gasoline in the US). This book concerns the upstream part of the industry and therefore how petroleum economics contributes to helping oil companies make decisions about searching for oil, producing it, and transporting it only as far as the point at which the downstream part takes ownership of it.

Where Is the Line Drawn? There are actually two lines I want to draw regarding what I believe is important to know and what isn’t: the starting one and the finishing one. On the finishing one first: I’m not out to boldly go where no petroleum economist has gone before (to coin a well-known phrase). On the contrary, I want to stay within the bounds of common petroleum economics practice and the tools of the trade that most petroleum economists around the world use every day; which are, without doubt, cash flow modeling and discounted cash flow (much more about these later). More exotic tools do exist, and I’ve seen some of them in action; and I’ve also witnessed first-hand why they usually fail, which they do for up to two reasons: • One is that too little attention is paid to basic concepts, basic assumptions, and basic calculations. In other words, someone has built a sophisticated house on foundations that can’t support it. The house can be as sophisticated as you like, but without proper foundations underpinning it, it will fall down. Now that issue can, of course, be addressed through education and training, and so perhaps it isn’t an insurmountable hurdle to overcome. • The other reason is less surmountable, however, and more tricky to overcome. It’s that senior managers who make decisions, who take ultimate responsibility for approving the multimillion-dollar investment before them, want to understand —no, more than that, insist on understanding—why the petroleum economist’s advice is what it is and how he or she reached that conclusion. And if the senior managers don’t understand, they won’t approve the investment; it’s as simple as that. It won’t do the petroleum economist any good waving a hot-off-the-press PhD thesis in front of them; he or she will quite likely be escorted from the board-room, and probably from the building as well. So, this book is restricted to the everyday tools of the trade that petroleum economists use, and anything less commonly used is beyond the line and out of bounds. As for the starting line, I don’t assume any knowledge whatsoever of cash flow modeling and discounted cash flow. Or any understanding of these individual words

even. All I assume is an interest in petroleum economics as I’ve just defined it.

How Am I Going to Say All This? Between these two lines—the starting one and the finishing one—is the contents of this book, and so let me say something about this space. This isn’t a report or an academic thesis. It’s a book; and one I’d very much like you to have some pleasure in reading. In the hope of achieving that aim, I include little in the way of mathematics. There will be technical language only where it’s completely unavoidable. And I also promise not to use the words “concept,” “framework,” “systematic,” and “methodology.” At the same time, however, I can’t promise you espionage, murder, romance, car chases, or alien invasions. Only the subject of petroleum economics explained as clearly, usefully, and interestingly as I can.

And a Few Last Comments The very last few housekeeping issues are these: When I use the word “oil,” I do so for convenience only and I mean any salable product (including, for example, gas) originating from a hydrocarbon-producing field. Second, terminology may differ among companies and even countries. I know, for example, that what a UK company calls a finance department, an American company calls an accounting group. I hope these differences cause minimal confusion. And finally, petroleum economists can, of course, be male or female; however, I have to use certain words now and again that prescribe them as one or the other, which I imply nothing whatsoever by doing.

To Summarize Petroleum economics is the voice that tells oil companies whether, for example, drilling an exploratory well, building a pipeline to connect an oil field to a refinery, or selling gas to a power station ought to make them richer or poorer. This book is about how petroleum economists provide that voice, which they most commonly do through the use of cash flow modeling and discounted cash flow.

Chapter 1

Introducing Cash Flow What petroleum economists care about above all else in the world is cash flow (in a professional, not a personal, sense that is; although many, myself included, consider it to be quite important in our personal lives too). So what is cash flow, and why do petroleum economists care so passionately about it?

1.1 Cash Flow Defined Cash flow is the flow of cash from someone to someone else, or from one organization to another one; that is, real money changing hands from a provider to a recipient. Providers can be just about anyone and just about any kind of organization, and so can recipients. And it comes in all shapes and sizes: it can matter or it may not, it may be expected or it may be a surprise, it may excite me or it may scare me, it may be in my control or it may not be. But whether or not it’s any of those things, the first thing we have to do is discover it. That may seem like an odd thing to say. In an oil company, I don’t have to look very hard to discover the cash flow involved when we sell a cargo of oil (obviously, we receive cash from whoever buys it), nor do I have to look very hard to discover the cash flow involved when we pay for a new pipeline (obviously, we pay cash to whoever builds it). I may not know exactly how much we’ll receive and how much we’ll spend, respectively, until the time comes, but I can discover the cash flows alright.

1.2 A Cash Flow Example Consider this example, then. Your boss asks you what cash flows another company (coincidently called Another Company Inc.) that owns a 10% stake in some oil fields in Russia experiences, which your company (coincidently called Your Company Inc.) owns the other 90% of. As shown in Fig. 1.1, the ownership structure (in black) and financing (in red) of the oil fields looks like this: Another Company owns a stake not in the Russian oil fields themselves but in a company (Let’s Be Friends Ltd.) that’s incorporated in the British Virgin Islands. Your Company is incorporated in the Cayman Islands and owns the other 90% of that company. The Russian oil fields are owned by a Russian company, Russia Oil Co., which is owned by a Cypriot company, Russia Holdings Ltd. Only Your Company has provided the necessary funding to the Russian company to explore for (and, we’ve been lucky, discover), develop, and produce the oil fields,

which it has done by (1) buying preferred stock in another Cayman Islands– incorporated company (Together We Stand Inc.), (2) lending money from there to a company incorporated in Luxembourg (Russia Capital Corporation Sarl), and (3) that company lending the money on to the Russian company. The oil fields are now healthily and happily producing and selling oil.

Fig. 1.1—Example of company ownership structure and internal cash flow.

What answer do you give your boss? It really isn’t obvious, is it? And the answer isn’t important, in fact; and nor does it matter if you don’t know what preferred stock is and have never seen a corporate structure like this before. My point is that cash flows aren’t always obvious. They can’t always be spotted without the need of even a moment’s thought. On the contrary, discovering cash flows requires investigation, proper investigation. And that investigation may have to be quite detective-like: talking to suspects you think might be involved, understanding motives, and finding out dates, facts, and figures. It may take you to your company’s finance department, or to engineers or tax experts, or to the project office, or to parts of the company you’ve never seen before and perhaps didn’t know even existed.

1.3 Why Is Cash Flow So Important? But what makes cash flow so important that petroleum economists are prepared to go to the very ends of the company to discover it? The answer is because establishing cash flows is the start of their journey to measuring whether their company should become richer or poorer as a result of doing something (drilling that exploratory well, for instance).

1.4 To Summarize Cash flow is the flow of cash from one person or organization to another. Sometimes its existence is obvious, but other times it isn’t and it has to be sought out. Discovering cash flows is essential however, because it holds the key to petroleum economists being able to answer the question that’s their responsibility to answer: should doing this (drilling this exploratory well, for example) make our company richer or poorer?

Chapter 2

What Is It About Cash Flow? So far, I’ve talked about cash flow in the very general sense of what it is and why it matters. Now I’m going to talk more specifically about the cash flows oil companies experience, which will occupy this chapter and the next two. Some of the cash flows oil companies expect to receive and some of them they expect to incur are cash flows they ought to “know,” although by “know” I don’t mean know for certain. Rather, they’re cash flows that people in the company with particular knowledge and expertise can estimate—for example, how much oil the oil field will produce and what price it will sell for (and so, between them, how much revenue the oil will generate for the company), as well as what it will cost to build everything to do with accessing it (platforms and pipelines perhaps). The next chapter is about that sort of cash flow. But some cash flows can’t be estimated like that. Some of them must be derived from separate—and often quite complex—calculations. These cash flows relate to a country’s “fiscal terms,” which is an important expression that warrants definition straightaway. “Fiscal,” according to my dictionary, is defined as “pertaining to the public treasury or revenue,” and “terms” conveys what “pertaining” means in practice. In other words, fiscal terms are taxes, although “tax” is actually far too blunt a word. Over the years, governments have invented ways of sharing in oil revenues that the word “tax” simply doesn’t do justice to. So a more helpful definition of “fiscal terms” is “what the government takes,” in all its varied and inventive forms. It’s the cash flow that goes to the government (or governments, plural, as we’ll see later). Chapter 4 is about that sort of cash flow and the calculations that derive it (I bet you can’t wait). Before any of that though, we should consider a few things that petroleum economists have to establish about cash flows first, however they’re derived and whatever they relate to.

2.1 Cash Flow: Beyond Discovery Petroleum economists need to answer three questions about a given cash flow: 1. Is it relevant? 2. How big is it? 3. When does it occur? We’ll consider these individually and in that order.

2.1.1 Is a Cash Flow Relevant? The number of cash flows that take place in even a small oil company can be many, never mind in a large international organization, in which the number is truly immense. Luckily, though, petroleum economists don’t have to care about all of them. In fact, they don’t have to care about many of them. They only have to care about cash flows that pass two tests. The first test is that cash flows must be brought about by the new activity the company is considering doing (I’ll continue using the same example: drilling an exploratory well). So these are cash flows that would newly come into the company or would newly exit it directly as a result of the company doing this new thing, whatever it may be. They may be brand-new cash flows that are created, or they may be changes to cash flows that already exist. Either way, the brand-new cash flows and the changes in already existing ones must be completely attributable to the new activity. The second test is best described with the aid of a picture (see Fig. 2.1). Here’s an oil company with a border around it. Once again, the company is shown as a corporate structure. Boxes represent companies that are owned by other companies higher up and ultimately are owned by the parent company at the very top. All companies can be depicted like this, and large international oil companies may have hundreds of boxes representing their various businesses doing various things in various parts of the world. Beyond the border is the rest of the world; that is everything that isn’t the company. To pass the second test, cash flows must cross this border; they must move from inside the company to outside it or from outside it into the company. Any cash flows that remain completely inside the border (and therefore completely inside the company)—for example, a cash flow from one box to another one—and any cash flows that remain completely outside it fail the test. A way to think of this is in terms of bank accounts. Cash flows that cross the border go from one of the company’s bank accounts (companies usually have more than one bank account) to someone else’s, usually one that belongs to another company or to the government. Or, of course: to one of the company’s bank accounts from someone else’s. Cash flows between bank accounts that both belong to the company don’t cross the border.

Fig. 2.1—Typical corporate structure of a company.

If a cash flow passes both tests, it qualifies as a “relevant” cash flow. Passing only one test, no matter how admirably, isn’t enough. It must pass both. Petroleum economists are interested only in relevant cash flows. If a cash flow isn’t relevant, then they aren’t interested in it and they can ignore it. 2.1.2 Cash Flow Example. An example, as ever, is helpful. Here’s the situation. An oil field of ours in Denmark has been producing oil successfully for several years. Now, our explorers have discovered another oil field nearby (as we speak, they are in the pub yet again, though this time with good reason: to celebrate). The new discovery can be produced through the existing field’s production facilities, although we’ll have to lay a pipeline back to them and also add extra equipment to them in order to do so. You are the petroleum economist and so must advise the company’s board of directors regarding whether to develop the discovery into a producing field (having determined if the company should become richer or poorer in doing so, although we’ve only just begun talking about how petroleum economists do that, so the board will have to wait a few more chapters yet). You start by thinking about what the relevant cash flows would be if we developed the field, as follows: i. The new field would produce oil that we’d sell to another company (one that owns a refinery, given that we don’t have one of our own). Is the money we’d receive for the oil a relevant cash flow? Yes it is, you conclude, because it crosses the border from outside the company into it; and if we don’t develop the field, we won’t otherwise produce the oil and sell it. So, the oil production and the incoming cash flow associated with it occur as a direct

result of our decision to develop the field. ii. But developing the new field would cost money. You’re told that we would need to drill three new wells, and those wells would be connected back to the existing field’s production facilities by a new pipeline, all of which work would be carried out by a contractor company that is completely separate from us. Regarding the extra equipment for the existing production facilities, however, it turns out that we acquired identical equipment last year for another project in Indonesia that won’t, after all, use it and is unable to sell it. “So,” the project manager tells you, “we shouldn’t have to pay full price for that because it’s secondhand now.” You think about that over a cup of coffee afterwards. “Hmmm,” you ponder. “The cost of the wells and pipeline is relevant cash flow. It crosses the border from inside the company to outside it, and it wouldn’t occur if we didn’t develop the new field. But if we already own the extra equipment and paid for it last year, then while our new Danish project may have to buy it from our Indonesian project to preserve the integrity of each project’s own accounts, that wouldn’t be relevant cash flow. True, cash flow may occur that wouldn’t have occurred had we not chosen to develop the new field in Denmark, but it doesn’t cross the border. It would be ‘from one box to another one,’ between bank accounts that both belong to the company. It would be different,” you ponder further, “if the equipment could be sold by the Indonesian project. In that case the Danish project is therefore depriving the company of that opportunity, which means the Danish project must assume responsibility for, effectively, costing the company money, real relevant cash flow money, the money it would otherwise have received from the equipment’s sale.” Economists call that an “opportunity cost” for the company and, in this scenario, it would count as a relevant cash flow, a cost, for the company’s Danish project. “And then there would additionally be relevant cash flow in transporting the equipment from Indonesia to Denmark and installing it there.” You finish your coffee and head back to the project manager’s office. iii. Not only would developing the new field cost money, so would producing it. There’d be costs to transport the oil to the place where we’d actually sell it, there’d be costs to ensure that the new wells and pipeline are maintained in good condition, and there’d be some additional costs associated with the existing production facilities, given the extra equipment it would have and the more oil production it would handle. Without going into any more detail than that, just for now let’s say you satisfy yourself that all these costs are relevant; they all cross the border, and they wouldn’t occur at all were it not for us developing and producing the new field. iv. If the new field made a profit, we’d have to pay tax. (We haven’t talked about tax yet. That’s for Chapter 4. So, until then, we won’t talk about it in any

great detail.) You visit the tax department, and the heads of the half dozen or so people who work there turn around as you walk past. You go to the head of tax’s (Mr. Smith’s) office and explain your task. Mr. Smith tells you that from a tax point of view, it should be straightforward, but he adds, “We’ll probably finance the development by lending money down from our Bermudan holding company. We can offset the interest we pay in Denmark against the tax we pay there. So you’ll need to take that loan and interest into account.” He then asks you, “Will you stay to have a cup of tea with us? We don’t get many visitors.” You say that you’re sorry, but you’re on your way to see the project manager and so, unfortunately, you don’t have time. He looks disappointed. As you leave, the whole department gathers to see you off, and they all wave as you disappear around the corner and out of sight. Glancing back, you think you can see a tear in Mr. Smith’s eye. As you walk back along the corridor, you think to yourself, “Tax is a relevant cash flow. It crosses the border, and we wouldn’t pay tax on the new field’s profits if we hadn’t developed the field in the first place.” Then you remember Mr. Smith also saying something about financing the project. “Hmmm,” you ponder once again. “But surely the loan isn’t a relevant cash flow? All that happens is cash moves from one box inside the company to another one (from our Bermudan holding company to our Danish company) when the money is lent, and then it moves back again in the form of interest and when it’s repaid. It doesn’t cross the border. That makes sense,” you decide, and you nod your head as you walk along. “But what about what Mr. Smith said about us offsetting the interest on the loan we pay in Denmark against the tax we pay there? What does that mean?” you ponder some more. “So what happens is the tax we pay in Denmark is reduced because those interest payments can be offset against our tax bill. So,” it dawns on you, “whilst the loan itself isn’t relevant cash flow, I do have to take into account how it affects the tax we pay, because the tax we pay is a relevant cash flow.” You walk on with a smile on your face and someone you pass asks you if you’re feeling alright because nobody ever comes away from the tax department smiling. 2.1.3 How Big Is the Cash Flow, and When Does it Occur? Of course, petroleum economists have to know the answer to the second question, how big the cash flows coming into and going out of the company would be as a result of it doing something new. Naturally it matters whether the sums involved are thousands of dollars, millions of dollars, tens of millions, hundreds of millions, or billions. But almost as necessary for petroleum economists to know is when these cash flows would occur, the answer to the third question. Briefly, because a whole chapter is going to talk about this later on, it’s why utility companies encourage us to pay our bills monthly rather than at the end of each 3-month period, and why “buy now, pay

next year” offers for new sofas appeal to so many people. But more about that later on. So where does all this talk about relevant cash flows and the size and timing of those cash flows leave us? The answer is in a very important place. It leaves us in the place where we know how our whole company’s cash flow profile would change from now into the future if it did the new thing. We’ve isolated what difference doing it would make to our company’s present and future cash flows. And that’s vitally important to know in our challenge to understand whether our company might become richer or poorer as a result of doing it.

2.2 To Summarize While there are many cash flows that take place in an oil company, petroleum economists care only about the ones that are relevant. Relevant cash flows are those that, first, are created by the company doing something new, such as drilling an exploratory well or developing an oil discovery into a producing field. These cash flows may be brand-new or they may be a change in an already existing cash flow. And second, they’re cash flows that cross the company’s border with the outside world: they come into the company from completely outside it or exit the company to completely outside it. Cash flows that don’t pass these two tests aren’t relevant cash flows and petroleum economists can ignore them. Petroleum economists also need to establish the size of relevant cash flows and determine when they occur (the latter of which we’ll talk much more about later). All this information together tells petroleum economists how their company’s overall cash flow profile from now into the future would change as a result of it doing the something that’s new, and it’s a vital stepping stone that allows them to begin to determine whether their company should become richer or poorer as a result.

Chapter 3

The Relevant Cash Flows Oil Companies “Know” This chapter is about the kind of relevant cash flows that experts in the company can estimate. Such cash flows are distinct from those that are derived from often quite complex calculations relating to countries’ fiscal terms—in other words, the cash flows that governments take from oil fields’ profits. The next chapter will cover that kind of relevant cash flow. Here, I’ll talk briefly about the relevant cash flows that are typical of each phase of an oil field’s life, which are: 1. Exploration—that is, the original search for an oil field. 2. Appraisal—that is, gaining an understanding of what has been discovered (of course, providing something has indeed been discovered) before deciding whether to develop it. 3. Development—that is, providing all the wells and production facilities so that a discovered field can produce its oil. 4. Production—that is, operating the wells and facilities safely and maintaining them in good order, ensuring the produced oil is fit and ready for sale, and transporting it to the place where it’s sold. 5. Decommissioning—that is, removing as much evidence as possible that an oil field was ever there and returning the locality to what it looked like beforehand. In no way is it my intention to explain how such cash flows are estimated. I couldn’t if I tried, and I take my hat off to those who do it. But petroleum economists mustn’t be ignorant of the process either. We use the numbers that these experts produce, which makes us their customers; and when a customer, it always pays to be an informed one. So in this chapter, simply making more informed customers of us regarding these cash flows is my hopeful goal. I ought to make a few obvious points first, however. Oil fields are found in very different places, come in very different shapes and sizes, and present very different challenges to make them work. So generalizing is difficult when we talk about them, but I don’t think that matters for this chapter’s general purpose. Also, the phases of an oilfield’s life aren’t necessarily as neatly sequential as I’ve just suggested. In practice, different phases sometimes merge together into one, and it isn’t uncommon for activities of previous phases, or even whole phases themselves, to be repeated later on in a field’s life, and sometimes more than once.

Here are the typical relevant cash flows that oil companies know.

3.1 Exploration 1. For onshore exploration, investigation of geological features on the surface that may indicate the presence of an oil field below it, for example, through fieldwork and satellite studies. (I don’t intend, by the way, to say much about the activities and processes I’m mentioning here that give rise to these cash flows. Once again, I couldn’t if I tried. But good books exist on each one of these subjects and, of course, courses too; so I delegate to them for that.) 2. Geophysical surveys that penetrate below the surface and indicate what may lie beneath it, for example, seismic, gravity, and magnetic. The preceding two areas of work are often referred to as “G&G,” for geology and geophysics. 3. Drilling an exploratory well, including all the services needed to support the operation (provision and transportation of supplies, materials, equipment, and crews for and to the drilling site); gathering information “downhole” that will indicate what’s there; collecting material, including cores (“plugs” of rock), from it; and testing (if a discovery is made) by producing from it for a short period of time. Note that not all these activities may be carried out for every well. 4. Study of what we can learn from the well by evaluating material recovered from it and processing information gathered during and after drilling. 5. Usually payment of annual fees to the government for the ongoing rights to explore over an area. 6. Office and onshore operations base costs, which I’ll talk more about later on.

3.2 Appraisal 7. Drilling an appraisal well or wells, commentary about which is similar to that for drilling an exploratory well (point 3). 8, 9, and 10. Same as points 4, 5, and 6. 11. Appraisal may involve producing from a well for an extended period of time to understand the field better than can be concluded from a short test. To do so can generate cash flows both out of and into the company, involving both the costs to collect, process, and transport the oil and the receipts from its sale.

3.3 Development

12. Drilling a development well or wells, commentary about which is similar to that for drilling of an exploratory well (point 3). 13. Study of what we can learn from the well (see point 4). 14. Preparation of an early engineering design to establish how to develop the field, to agree on the specifications of plant and equipment that will be needed, and to better understand the project’s execution plan and what it might cost to fully design, build, install, and commission it. 15. Creation of a detailed design that can be given to manufacturers to build. 16. Manufacture of everything necessary to receive production from the well(s) and deliver salable products. 17. Connecting all the manufactured parts together and making sure the whole system works as expected. 18 and 19. Payment of fees and costs (see points 5 and 6).

3.4 Production 20. The day-to-day cost of the whole production operation, including paying the people who make it work, getting them to and from the field, and looking after them when they’re there; fuel and power for all systems of the production operation; provision of supplies and their transportation to and from the field; and routine maintenance. 21. Chemical treatments, especially in pipelines to maintain the flow of production and prevent the buildup of substances that might hinder it and also to preserve the condition of the pipelines. 22. Wells, which can fail either partially or completely for a number of reasons, require maintenance that may, depending on circumstances, be costly. 23. Plant and equipment must be regularly inspected and maintained to keep it in good working order and preserve a safe working environment. 24. Oil may be transported through systems that aren’t owned by the field that produced it (pipelines, for example) and may also be subjected to further processing. Owners of those systems charge a fee—a “tariff,” usually per unit of volume—for providing those services. 25 and 26. Payment of fees and costs (see points 5 and 6). 27. Receipts into the company from the sale of oil. 28. Just as we may pay tariffs to transport and process our oil in other fields’ systems, other fields may pay tariffs to us to transport and process their oil in ours. 29. Royalties (sums of money) that oil companies have agreed to pay other oil companies and individuals for a variety of reasons, which I’ll expand on in the next chapter.

3.5 Decommissioning

30. The cost of taking the field’s wells out of service. 31. The cost of taking away all plant and equipment and returning the environment to a safe, clean, and natural condition. 31 and 32. Payment of fees and costs (see points 5 and 6). 33. Possibly the sale of certain plant and equipment to another company for further use or for scrap.

3.6 To Summarize Here’s a diagram (Fig. 3.1) by way of a summary of all these cash flows.

Fig. 3.1—Summary of cash flows.

Chapter 4

The Relevant Cash Flows Oil Companies Have to Calculate This chapter is about the kind of relevant cash flows oil companies calculate as opposed to the kind they estimate (which were discussed in the last chapter). So, basically, I’m going to talk about tax and the various other forms of income that countries’ governments receive from oil fields and oil companies. But here’s a short story first. I always imagined tax was such a complex subject that you had to have a PhD in physics, or something like it, to understand it; and that was especially a concern of mine in my early years as a petroleum economist because tax involves such large sums of money. Well, one day back then, I didn’t understand some calculations relating to tax, and so I wandered along to the tax department for some help. There I met a young gentleman who, it turned out, hadn’t been with the company for long either. Before I sought his help, we chatted for a few minutes. “So what’s your background,” I asked. “I’ve just finished my PhD in physics,” he replied. True story. But don’t be afraid. There’s nothing to be frightened of, I promise. As long as we stick together and stay alert, we have nothing to fear. This chapter is divided into four parts. The first part covers fiscal terms in the country in which an oil field resides (as a reminder: fiscal terms are simply the rules that determine what the government earns from an oil field or oil company). The second part talks about taxes when profits generated by an oil field in one country leave that country for another one (which happens all the time in international oil companies). The third part covers how that money is treated by the other country’s tax authorities when it gets there. And the fourth part covers indirect taxes, the meaning of which I’ll explain when (and I’m confident it’s “when” and not “if”) we get there. This brings me to a plain and simple fact however: I can’t in the space of just several pages provide a detailed explanation of what is a fantastically complex subject. To do so would be an impossible undertaking and my ambition for this chapter is much more modest than that. Every country’s fiscal terms vary around some common, basic principles, and all I aim to do is explain what those common, basic principles are so that you can go and talk intelligently to your own tax department about your project. They can then tell you if your project varies from those principles and, if so, how, and describe the applicable fiscal terms with more precision.

Petroleum economists aren’t tax experts, and because of that, and also because taxes are such important and often complex cash flows in petroleum economics, going to talk to their tax departments is something that petroleum economists simply have to do—and they have to do it a lot. So, here we go with just one last piece of advice: stay calm at all times. If you feel yourself starting to panic at any point, stop reading and breathe deeply for a few moments; and if that doesn’t work, take a break and make a cup of tea before continuing.

4.1 Fiscal Terms in the Country in Which an Oil Field Resides Tax authorities in most countries have one set of fiscal terms for oil companies and another set for other, non-oil companies. They generally treat oil companies differently. They may use a completely different approach to what most of us might immediately understand by the word “tax” (as much of Africa and Asia do, for example, in the form of production sharing contracts, which we’ll talk about later), or they may just charge oil companies extra tax over and above what every company already pays (which Norway, Denmark, Russia, and Australia do, for example). I’m not going to discuss the rights and wrongs (in my humble opinion) of oil companies’ differential treatment by countries’ tax authorities, nor about the relative merits of how they achieve it, because countries’ fiscal terms are, generally, what they are and it’s unusual that petroleum economists can design or change them. We can moan that they’re too severe or that they don’t make sense, but usually our opinions don’t matter. So I’d much rather spend our limited time together simply talking about how they work. I alluded just now to there being two broad approaches to fiscal terms that countries’ tax authorities apply to oil companies: • An extension of the ones that apply to every company • A completely different approach to the ones that apply to other, non-oil companies It’s this distinction that produces the two broad types of fiscal terms that are used across the oil industry: “concession,” also called “tax/royalty,” which is applied in countries that fit the first bullet point, where some more tax is added to the tax that every company already pays; and “contractual,” which is used in countries that fit the second bullet point, where a very different approach is taken. Some countries use both types of fiscal terms, although only one type ever applies to an individual oil field save in just one or two instances in the world. The next few pages talk about these two broad types of fiscal terms one at a time, although it might be helpful to describe them briefly first, just to illustrate the fundamental difference between them. And from now on in this chapter I’ll boldface certain words that I think would benefit from explaining, which I’ll do as soon as possible after I’ve used them for the first time.

4.1.1 The Difference Between Concession and Contractual Fiscal Terms. From a petroleum economist’s point of view, the difference between concession and contractual fiscal terms lies in how the profits an oil company makes from an oil field are determined for the purpose of deciding how much of it the oil company keeps and how much the government takes. Take a producing oil field. In a concession world, the oil company that owns the oil field spends money finding it, developing it, and producing it; it sells the oil the field produces and pays tax on any profits it makes from the whole venture. Regarding the word “owns”: strictly speaking, oil companies don’t own oil fields. Ownership usually rests with countries’ governments, which award oil companies the rights to explore for, produce, and sell oil from one or more of them for a period of time in return for assurances that they’ll honour any promises they’ve made (for example, to drill a certain number of exploratory wells) and abide by rules regarding their conduct. Not always, however; onshore in the US, for example, ownership rests with individual landowners. So when I use the word “owns” in this context, I use it loosely. Back to the fundamental difference between concession and contractual fiscal terms: in a contractual world, however, the oil that an oil field produces is divided up between the oil company and the government according to a procedure that’s set out in a contract between them. In both cases the oil company makes a profit, it hopes, but the profit is just calculated differently; and that profit—whichever approach is used—can be taxed. In that respect, therefore, it’s a shame that concession fiscal terms were ever also called “tax/royalty” because tax doesn’t apply only to that approach; but they are and that’s that. 4.1.2 Concession, or Tax/Royalty, Fiscal Terms. As its other (unfortunate) name suggests, concession fiscal terms have two parts: a tax part and a royalty part, although they don’t necessarily include both; some concession fiscal terms have only a tax part. Royalty. We’ll begin with royalty because it comes first in computations of concession fiscal terms. Royalty is a proportion of an oil field’s oil that the government takes as its own, without paying for it. The payment may be provided by oil companies “in kind” (meaning in actual oil) or “in cash” (meaning money; the cash equivalent of the relevant amount of oil), or more commonly as either and the government chooses which method of payment it wants when a royalty payment is made. The payment is invariably defined as a proportion and not a fixed amount of oil production, so it would be, say, 5%, rather than, say, 100,000 barrels in a year. It’s a proportional slice of whatever’s produced. The amount typically ranges from 5 to 15% and can apply to any product the field produces. That amount goes straight to the government authorities. It belongs to

them from the start and the oil company has no claim whatsoever to the income from it when it’s sold. In Monopoly speak: it goes straight to the government; it does not pass go and does not collect two hundred pounds (ignore this sentence if you’re not familiar with the game of Monopoly). In practice, governments like royalty and oil companies don’t. Oil companies first: like anyone else, oil companies prefer to be taxed (if they have to be taxed at all, that is) on the profits they make. But oil fields pay royalty whether they’re profitable or not. So long as they produce oil, a portion of it is taken by the government. Even if an oil field makes no profit at all, the oil company owning it still pays royalty. In particular, royalty hurts oil companies in the early years of an oil field’s life when they’ve just spent a great deal of money drilling wells and building platforms and so on, which they’re eager to recoup as fast as they can; yet as soon as an oil field begins producing, royalty is immediately due. Governments, on the other hand, like royalty, and pretty much for the same reasons that oil companies don’t like it. They have the certainty that they’ll receive income from an oil field whether or not it’s profitable for the oil company; and not only that, they’ll receive it straightaway, as soon as production starts. So what’s the difference between royalty and tax? (That isn’t a joke by the way, like what’s the difference between a tax accountant and a coconut? Answer: you can get a drink out of a coconut.) For petroleum economists, there isn’t one really, except in the nature of their calculations. Apart from that, the difference between them may boil down to nothing more than the way they’re both reported in the company’s accounts, but that isn’t important here. Before we leave this section, it’s worth noting that the word “royalty” is also used to describe payments that oil companies sometimes make to, and receive from, other companies or even private individuals. Royalty of this sort may come about because one oil company has sold an unexplored area to another one and they’ve agreed that should oil be found there by the buyer, the buyer will pay the seller a “royalty”—a fee —when that oil is produced. It may also result when an oil company makes use of property owned by a private individual, who is then compensated in the form of a “royalty.” Although these so-called royalties represent relevant cash flows, they aren’t a part of a country’s fiscal terms. Tax. There are two kinds of tax that I’ll talk about next: production taxes and profit-based taxes, and I’ll talk about them in that order. Production Taxes. Of course, strictly speaking, all taxes on oil companies—at least in relation to their ownership of oil fields—are production taxes because it’s producing oil that earns oil companies the profits they pay tax on. Here, though, by production taxes I mean taxes that are based on the oil an oil field produces rather than on any measure of the profit it makes. In that respect, these taxes are therefore quite royalty-like in the way they work,

although they differ from royalty in their mechanics. Whereas royalty is a slice of whatever oil is produced, production taxes are a cost per barrel, ton, or other measure of oil produced, although the cost may be adjusted by factors that increase or reduce it, taking into account, perhaps, the oil field’s size or location or other more or less challenging circumstances that make it more or less profitable. And that means that even taxes as relatively straightforward as these follow no common algorithm in the countries that use them. Despite the fact that they can be made more or less sensitive to oil companies’ fortunes than royalty, at the end of the day they still bear the hallmarks of royalty (by not being profit based) however, and so they have similar advantages and disadvantages in the eyes of oil companies and governments. Profit-Based Taxes. Whereas not all concession countries have a royalty or production taxes, they all have some kind of profit-based taxation. “Profit” is a word we use all the time, of course: we make a profit when we sell something for more than we paid for it and we sustain a loss when we sell something for less than we paid for it. About the most important feature of profit-based taxation is that we pay tax when we make a profit and we don’t pay tax when we don’t make one. It’s important enough to say again: if we make money, we pay tax; if we lose money, we don’t. A former tax teacher of mine (I say “tax teacher”; the company I worked for thought of him more as someone they employed to manage their tax department) always used to say that paying tax isn’t a bad thing; at least it means you’re making money. Profit-based taxes work like this: 1. We take the value of what we’ve received. 2. We subtract, in the form of deductions, what it has cost us to create that value. 3. We multiply what’s left, which is our taxable income, by a tax rate; and the answer is the current tax that we have to pay. Again, I’ll spend a moment on each of these terms. Value. The value of what we’ve received is what everything we’ve received is worth in the eyes of the tax authorities. The “in the eyes of the tax authorities” part is important. Let’s say we sell a barrel of oil on the open market for USD 92.74. The tax authorities would agree that it’s worth USD 92.74 because of the very fact that we sold it on the open market. That’s what the open market was prepared to pay us for the barrel of oil, and so that must be its fair value. If we sell the oil cheaply, as a favour to a friend, say, the tax authorities’ alarm bells would ring, concerned that they’re missing out on some tax revenues; which may result in their assuming, for tax purposes only, that we did sell the barrel for USD 92.74, or a price much more like it. This is a concern that some countries circumvent completely by applying an ‘approved

price’ to any barrel of oil that’s sold (Norway does, for example, with their “norm price”) which is a price set by the tax authorities that producers must use for tax purposes. Another Scenario. Let’s say we agree with the owners of another oil field that we’ll transport their oil in our pipeline and in return they’ll give us 5% of the oil we transport for them. Do we pay tax on the oil we receive? Most certainly we do. Paying us in oil is still paying us, and so we have to calculate the oil’s equivalent value, and that’s the amount we declare to the tax authorities that we’ve received. Usually, we’ve had to spend some money in order to create value, and profitbased taxation allows us to subtract the money we’ve spent from the value of what we’ve received in order to arrive at the amount we have to pay tax on. But what we subtract isn’t generally as simple as what we’ve spent, when we’ve spent it. What we subtract are deductions, although, strictly speaking, I ought to say “allowable deductions” because whatever it is we’ve spent money on, we can deduct only what we’re allowed to, what the tax authorities say we can, which are the legitimate costs of being an oil company and carrying on an oil company’s business; no more, no less. Deductions. Deductions are what we’ve spent but not necessarily when we’ve spent it. When the tax authorities let us deduct what we’ve spent depends on what we’ve spent the money on. We can deduct certain amounts we’ve spent all at once, straightaway, if we’ve spent them on the kind of things that we buy, use immediately, and then they’re gone, lost, and we don’t have the benefit of them anymore. Examples are salaries, consumables that are immediately used up, the rent of an office perhaps, interest on a loan, and oil transportation costs when we use someone else’s pipeline and pay a tariff (a transportation fee) in return. There are other amounts that we can’t deduct all at once, straightaway, however, and we have to deduct them a bit at a time over several years; and, what’s more, usually starting only when we actually begin using whatever it is we’ve bought. That would be the case if we’ve spent the amounts on the kind of things that we buy and keep on using year after year. Examples are wells, platforms, and pipelines (our own pipelines, the ones that we own, that is). The former sort of costs (of things we buy, use immediately, and then they’re gone) are called “operating costs,” or “revenue costs” (because they’re usually met out of the revenue that a company earns), and the latter sort (things we buy and carry on using year after year) are called “capital costs.” Operating costs are the costs that are deducted all at once, straightaway, and capital costs are the ones that are deducted a bit at a time over several years, usually starting only when we begin using whatever it is we’ve bought—all of which translated into tax speak is that operating costs are expensed and capital costs are depreciated, usually starting only when the asset is placed into service (“the asset” being the well, platform, or pipeline, for example). So “expensed” means deducted all at once, straightaway, and “depreciated” means deducted a bit at a time over several years. “Expensed,” therefore, needs no further definition; “depreciated,” however, does.

Individual countries explain how they apply depreciation. It may be a fixed proportion each year over a number of years of the total amount we’ve spent: so a capital cost of USD 100 million might be depreciated (and deducted) as USD 10 million per year over 10 years; which is called “straight-line” depreciation (because USD 10 million per year over 10 years looks like a straight line on a graph that shows deductions on the y-axis and years along the x-axis). Or it may mean a proportion of whatever’s still left to be depreciated each year—say, 25% per year: so a capital cost of USD 100 million would be depreciated as USD 25 million in the first year (25% of USD 100 million, leaving USD 75 million still to be depreciated), then USD 18.75 million in the second year (25% of USD 75 million, leaving USD 52.25 million still to be depreciated), and so on; which is called “declining-balance” or “reducing-balance” depreciation. There are several other depreciation methods, some of which are derivatives of the preceding two and some of which are different, and I don’t intend to talk about all of them. Suffice it to say that countries’ tax authorities make clear what method oil companies must use. But be aware that different methods may be used for different kinds of things or parts of assets (different parts of a well or production facility, for example)—and your tax department will know, or can find out, the rules that apply in a given country. The moment an asset is placed into service is when it actually starts working, not when it’s manufactured, nor even when it’s installed, and it may sound trivial to make the distinction. So we have to wait a year or two after we pay for the asset before we can start depreciating it and claiming deductions. It’s a nuisance, but it isn’t a big deal. Or is it? Another Story. I was the petroleum economist for a potential acquisition of a small additional interest in an oil field in which the company I worked for already owned a significant interest. So the tax calculations for the oil field already existed in the company. However, I noticed when I looked at them that deductions for capital costs were starting immediately when the costs were incurred rather than, as they should have been, starting only when whatever had been bought was placed into service. The country in which the oil field resided had two profit-based taxes: a fairly standard one, but also another potentially quite brutal one that was, however, triggered only when certain quite rare circumstances occurred. These circumstances were when how much income an oil field had earned divided by how much money had been spent developing and producing it reached a certain trigger value. Both parts of that ratio were defined specifically—and somewhat curiously I might add, but I won’t go into that. Our calculations as they were being done (deductions relating to capital costs starting immediately when the costs were incurred) showed the trigger value not to have been reached yet and, in fact, never being reached in the oil field’s lifetime. However, when the calculations were corrected, they showed that it would be reached in the following year and the only way to prevent that happening was to acquire the small additional interest that was for sale (because most of the acquisition

cost could be added to the denominator of the ratio, thereby decreasing it and moving it away from the trigger point). So we did that, and had we not done so, my company would have paid an extra USD 1 billion in taxes over the oil field’s remaining lifetime. The moral to the story is that when it comes to tax, never prejudge what may or may not be important. But let’s get back on track. Operating costs and capital costs are common language in oil companies. You hear them all the time. But sometimes classifying costs as either one or the other and then deducting them according to the above general rule (that operating costs are deducted immediately and capital costs are deducted over time) is risky. Some costs may be difficult to classify, and some you may think are clearly one or the other but aren’t according to a particular country’s rules. An example is drilling, which some countries treat entirely as a capital cost, whereas others treat it as part capital cost and part operating cost. So it really is advisable, at least if and when in doubt, to let your tax department tell you how a cost is treated in the context of a tax deduction: whether it’s expensed like an operating cost or depreciated like a capital cost (and if so how), or a combination of the two. One deduction worth special mention is the costs associated with decommissioning an oil field. Decommissioning means ceasing the oil field’s production, “plugging and abandoning” its wells so that they can’t produce anymore, removing all its production facilities and returning the locality to its former, natural condition. Although at least some of the cost relating to its wells and probably all of it relating to its production facilities would have been depreciated for tax when the oil field came into being, the cost of decommissioning it all is expensed for tax (at least, I don’t know of any country that applies concession fiscal terms where it isn’t). Now you may be thinking, “Wait a minute, it’s all very well letting me deduct my decommissioning costs, but I’m not producing anything anymore, so I don’t have any revenue from selling oil to deduct them against.” And you’d be right. Good thinking. Concession fiscal terms deal with that in two ways, both of which I’ll explain in a minute. Before I do that, the last point I want to make about deductions is that where they relate to capital costs that are depreciated, they aren’t cash flow anymore. We’ve taken a cash flow—a cost—and transformed it into something else, so it’s no longer an amount of money we pay someone; it’s now an artificial stream of numbers that’s been created for the sole purpose of calculating tax. It’s become, and is called, a “non-cash cost.” Now, carrying on with our definitions from earlier (which I hope isn’t starting to seem like a lifetime ago). Taxable income is the difference between the value of what we’ve received and the deductions we can subtract from it. Needless to say, if the value of what we’ve received during the year is greater than the deductions we can claim, then our taxable income is positive and we’ll have some

tax to pay. (Tax is usually assessed over a period of a year, although it isn’t uncommon that interim payments are made, say, every quarter.) If, however, it’s the other way around and the deductions we can claim in the year are greater than the value of what we’ve received, then our taxable income is negative and we don’t pay any tax. In that case, our losses (remember, we haven’t necessarily lost real money; these are losses only in the context of establishing whether we have to pay tax this year) can usually be carried forward into future years. Then, they become an extra deduction in the following year, and our taxable income in that following year is therefore less than it would otherwise have been, and so is the tax that we’ll have to pay in that year. And if, in that following year, our taxable income is negative again, then our losses are carried forward once again, and so it goes on (though not necessarily forever; note the paragraph after next). Some countries also allow these losses to be carried back, meaning brought back in time to earlier tax periods. Previous years’ taxes are then revisited and recalculated, resulting in repayments being made by the tax authorities for, in this case, overpaid taxes in earlier years. And this is one way that concession fiscal terms deal with the decommissioning costs mentioned a little while ago. Those costs may be able to be carried back to earlier tax periods, and so we can still deduct them, but against earlier years’ taxable income when the oil field was still producing. Not all concession fiscal terms allow this to happen, however; and in that case, there’s only one hope for deducting decommissioning costs—which is coming up in another minute. The rules about what happens in the case of negative taxable income differ, as ever, among countries and, like most things relating to tax, can be complicated. Countries may apply a limit to how long a loss can be carried forward—and if the loss hasn’t been used by then, it’s lost—or on how far in time it can be carried back. In the spirit of the aim of this chapter, further information is available from your local, friendly tax department. The tax rate is simply the proportion of taxable income that is paid in tax, so it could be 20% or 40% or whatever it happens to be in the country we’re in. The tax that results from multiplying a company’s taxable income by the applicable tax rate is called its current tax liability. Current tax is the tax an oil company owes the government from a given tax period’s activity as a result of what it’s sold, what it’s spent, and anything else that has contributed to taxable income. Cash tax, on the other hand, is the tax that’s actually paid during a period of time, which may not be the same as the tax that’s calculated as owed for that same period. Often, some of the owed (current) tax is paid in the same tax period and the rest is paid in the following one, after all the relevant information has been collected and the calculations have been done so that it’s clear what the tax bill for the period actually was. It’s cash tax, therefore, rather than current tax that represents relevant cash flow. The last question I’m going to address in this section is to what do profit-based taxes apply? Is it an individual oil field, is it a whole company (that might own several oil fields), or is it something else?

Again, I’m afraid, there isn’t a clear answer. Some taxes apply to an individual oil field, and some apply to as many oil fields all together that a company owns in a country, as if they were just one big oil field. The term tax people use when they talk about this is a “ring-fence.” A ring-fence defines the scope of the tax calculation— how many oil fields it includes—and is so called because you can draw an imaginary ring-fence around one or more oil fields and then treat whatever’s inside it in aggregate, all added together, for calculating tax. If a tax has a ring-fence around an individual oil field, then it applies just to that oil field and that one alone. If the ring-fence is around, say, the whole country, then the tax applies to all the oil fields a company owns in the country lumped together, although instead of “lumped together” tax people say “consolidated” because they like to use long words and therefore sound more clever. Even where a ring-fence is around a whole country, however, for oil companies it usually also exists only around their upstream activities (that are connected with oil exploration and production) so that those activities can’t be consolidated with anything else the company does. In other words, oil companies’ upstream activities are usually kept separate for tax purposes. I’m sure I don’t need to say where to go for information about what ring-fence applies to your tax calculations. Consolidation is also the second way that concession fiscal terms deal with the decommissioning costs mentioned a little while ago. If a tax is consolidated at the country level and there are other oil fields inside the ring-fence generating taxable income between them that’s greater than the decommissioning costs of the oil field at the end of its life, then those decommissioning costs can be deducted there and then. All that happens is that those costs reduce the company’s taxable income for that tax period and so it pays less tax. In exactly the same way, consolidation is also helpful at the other end of an oil field’s life, especially for exploration and appraisal costs that are expensed. Again, if a tax is consolidated at the country level and if other oil fields inside the ring-fence are generating enough taxable income, then those costs can be deducted immediately, and certainly much sooner than if their deduction has to wait for the new oil field to generate taxable income of its own (which could be up to several years later). Achieving deductions for tax is a good thing because when we deduct costs, we pay less tax. Achieving deductions as soon as possible is also a good thing because we pay less tax sooner rather than later: we pay less tax today and more tomorrow rather than the other way around. I’ve mentioned this idea once already, at the end of Chapter 2. To paraphrase what I said then: when cash flows occur is important for petroleum economists to know. We place more value on cash flows occurring sooner than on those occurring later, which is why utility companies encourage us to pay our bills monthly rather than at the end of each three-month period, and why “buy now, pay next year” offers for new sofas appeal to people. That’s the whole subject of the next chapter. And that’s all I’m going to say about concession fiscal terms.

So far so good. For those of you for whom numbers speak louder than words, Appendix 4.1 shows two worked examples of everything covered in this section: one that assumes an individual oil field isn’t consolidated with other oil fields and another that assumes it is (and that those other oil fields always generate more positive taxable income than the individual field generates negative taxable income). I encourage you to replicate these examples in a spreadsheet of your own. Those of you for whom numbers do no such thing, by all means ignore these examples and move on. And now I’m going to talk about contractual fiscal terms, which are rather different than concession ones. Very different, in fact. 4.1.3 Contractual Fiscal Terms. There are two types of contractual fiscal terms that I’ll discuss here. The most common is the production sharing contract or production sharing agreement or production sharing arrangement (they’re all the same thing; PSC or PSA for short), which is common in certain parts of the world, including Central and Southeast Asia and parts of Africa. Less common are service contracts or technical service contracts (they’re also the same thing, and again they may be called an agreement rather than a contract), which, from a petroleum economist’s point of view, have much in common with PSCs. So I’ll spend most of this section talking about the more common PSCs and then say something about service contracts at the end. Production Sharing Contracts. The term “production sharing contract” refers more to a family of contracts than to a single, universal one. All PSCs belong to the same family—they’re all the same in a broad sense—but they all have different characteristics and features in the countries that use them. In a nutshell, a PSC is a contract between one or more oil companies on the one hand and the government authorities of a country on the other that describes a program of work to explore for, develop, and produce oil fields (if any are discovered) or to redevelop and produce already discovered oil fields within a designated area. It also describes how the program of work will be organized and managed, who out of the oil companies and the government authorities is responsible for what, and what they will both receive in remuneration. Production sharing contracts have a life of 25 years or so, although contracts can be terminated early or extended in certain circumstances, they are about 100 pages long and contain no jokes or funny cartoons. The one or more oil companies are designated “the Contractor,” which is slightly confusing when, generally in the oil industry, the term “contractor” is also used for companies that oil companies employ to build or install something for them, a platform or a pipeline, for example. Where there’s more than one oil company, each one is a proportionate member of the Contractor, and then whatever the Contractor is obliged to pay or is entitled to

receive under the terms of the PSC is divided up in those proportions. So, say, two oil companies make up the Contractor in a PSC, one with a 70% interest and the other with a 30% interest, anything the Contractor is obliged to pay is provided by the oil companies in those proportions, and anything the Contractor is entitled to receive is allocated to them in those proportions. The Contractor finances the program of work; so all the exploration and, if oil is discovered, development and production costs are paid for by the Contractor. The government authorities don’t pay anything toward these costs (except if their country has a national oil company that forms part of the Contractor group of companies). If the Contractor comprises more than one oil company, then one of them is designated as the operator, in the same way that any oil field has an “operator” who manages things. However, in a PSC world, one extra job the operator has over and above its normal duties is to file a PSC return with the government authorities on behalf of all the oil companies that make up the Contractor. A PSC return is usually filed annually, but it may be, say, quarterly, and is a record of how much oil has been produced since the last return, how much revenue has been earned, and what the results are of applying the contract’s remuneration mechanisms (among other information) for that period. This brings me nicely to those aspects of PSCs that interest petroleum economists the most. There are usually a half-dozen or so pages in a PSC that describe the contract’s remuneration mechanisms (although I would always but always advise petroleum economists to read the whole contract because it isn’t uncommon that matters relevant to them are tucked away in other sections). These crucial pages can be found in a section called something like “Recovery of Petroleum Costs and Production Sharing,” or words to that effect. Now, what follows is perhaps the most important statement in these few pages: what an oil company receives in revenue in a PSC world isn’t its share of the number of barrels of oil that an oil field produces multiplied by the oil price per barrel it sells them for as in a concession world, it’s the sum of something called “cost oil” and something called “profit oil.” Cost oil pays the Contractor back in oil for the money he’s spent. (Remember that the Contractor, not the government, finances the program of work.) Production sharing contracts therefore provide the Contractor with cost recovery, which is the principle of paying him back what he’s spent. (Referring to the Contractor as “he” and “him” doesn’t sound quite right, but never mind.) Needless to say, the detail of exactly how they do that isn’t common to all PSCs, and they all go about it slightly differently. What is common to them, however, is an account that tracks how much money the Contractor spends and how much he’s owed at any given time, called the cost pool. Cost pools work like this: when a Contractor—or the oil company or companies making up the Contractor—spends money, that money is added to the PSC’s cost pool. The contract specifies which costs are admissible, and the government

authorities keep a very careful eye on Contractors to ensure that they add no more than those costs. (In practice, however, disputes can still arise between the Contractor and the government authorities regarding the admissibility of costs which result in their having to engage in discussions to resolve them.) Often, exploration and appraisal costs are added in full and immediately, as are operating costs, but capital costs are drip-fed into the cost pool, starting only when the assets associated with them are placed into service, much like how capital is deducted in profit-based taxes in concession fiscal terms. So, for example, capital spent on an asset that will become operational in the following year might be added to the cost pool in equal amounts over a 5-year period, starting the next year. Decommissioning costs may or may not be addressed. Given that PSCs invariably stipulate that once costs are recovered, any assets acquired for an oil field (that is, all the wells plus production, processing, and other equipment) become the property of the government, some contracts are silent on the matter of their decommissioning, and it’s implicitly understood—although rarely explicitly stated—that the costs of decommissioning are therefore the government’s responsibility. In contracts that do address these costs and specify that they’re the Contractor’s responsibility to pay, they’re added to the cost pool during an oil field’s life before decommissioning comes about. Estimates of decommissioning costs are made, and those cost estimates are added to the cost pool usually spread over the oil field’s life, or perhaps only the last several years of it. So the costs aren’t so much “recovered” (because they haven’t been spent yet) as “provided in advance” so that when decommissioning does come about, the costs of it have already been recovered. Then in each reporting period—each year, say—oil is allocated to the Contractor out of what has been produced that is equal in value to the contents of the cost pool. The cost pool’s contents are then cleared, and the cycle continues: more costs are added (in the following reporting period), oil production of that value is allocated to the Contractor, and the cost pool’s contents are cleared. And the cycle starts again. If not enough oil has been produced to cover the cost pool’s entire contents, then what costs aren’t recovered are carried forward into the next reporting period and recovered then. And if the costs aren’t recovered then either, they continue to be carried forward until they are recovered. If this sounds familiar, it is; it resembles the carrying forward of taxable losses in profit-based taxes in concession fiscal terms. Sometimes, as an alternative to drip-feeding capital costs into the cost pool, they’re added in full. But then, usually, only a proportion of the oil that has been produced is allocated for the Contractor to recover his costs. So, say, only 50% of production in a reporting period is allocated, in which case it’s said that the PSC has a cost recovery ceiling—that is, a ceiling limiting how much oil in a reporting period is allocated to cost recovery for the Contractor. And again, if that isn’t enough to cover the entire contents of the cost pool, then what isn’t recovered is carried forward to the next reporting period (and then to the next one if necessary, and so on). In the event that the proportion of oil production allocated to cost recovery more than covers the cost pool’s contents, then what’s left over—which is often called “excess cost

oil”—is treated similarly to profit oil (an explanation of which is coming up). The rest of the oil produced in the reporting period—after the Contractor’s cost oil and, if applicable, excess cost oil are accounted for—is profit oil. Profit oil is shared between the Contractor and the government authorities, and the PSC explains how. It might be a straight percentage split between them, or it might be something much more complicated depending, for example, on production rates from the oil field or on how much revenue the oil field has generated relative to how much money has been spent developing it. (I don’t propose to explain each and every possibility because the variations are too many.) Then, the sum of the Contractor’s cost oil and its share of profit oil is the Contractor’s remuneration entitlement from the PSC. That’s what the Contractor gets back. It’s the revenue the Contractor receives, in the form of oil, which is divided up between the oil companies that make up the Contractor if there’s more than one of them, and sold. Cost oil and profit oil generally apply to the PSC as a whole and not, where there are several fields in the area covered by the contract, to individual fields. To borrow some expressions from the concession world: usually, a PSC’s geographical boundary represents a ring-fence, inside which oil fields are consolidated. So all the revenue earned by the oil fields within that boundary and all the costs incurred on them are aggregated for the purpose of calculating the Contractor’s cost oil and profit oil. There are a few occasions, however, unfortunately, where cost pools apply to individual fields in the PSC—and I say “unfortunately” because that can add significantly to the length of the computations required to model the contract. And in essence, there you have it. That’s how production sharing contracts work. The oil produced by an oil field is divided into cost oil, which repays the Contractor’s investment in it, and profit oil, which is shared out in some way between the Contractor and the government authorities. You won’t be surprised to hear that in practice PSCs aren’t as simple as that, however. It won’t have escaped your notice that words like “usually,” “often,” and “generally” have been liberally scattered over the last few pages. That’s because beneath the essence of these contracts, they differ a lot—not only in the nature of their cost-oil and profit-oil calculations but also regarding other features they may or may not include. Some, for example, might also have a royalty (which disappears off to the government authorities before cost oil and profit oil are even contemplated); some might also include one or more other costs that the Contractor has to bear, which aren’t necessarily even connected with oil fields and sometimes aren’t recoverable either (for example, relating to social and educational programs in the country). There may also be sums the Contractor has to pay when production reaches certain milestones, either daily rates or cumulative volumes, called “bonuses,” although note that these are bonuses paid by the Contractor to the government authorities, not the other way around. And they might also include tax, which a few words need to be said about separately. Tax here is tax that Contractors pay on the profits they make, so we’re talking

about profit-based tax. (Warning: you may find the following sentence confusing.) Sometimes it’s tax that Contractors are liable for and pay, and sometimes it’s tax that they’re liable for but the government pays for them. Let’s take the first, less confusing part of that sentence first. In this case, tax is calculated in exactly the same way as the profit-based tax discussed earlier (although, on occasions, PSCs are silent regarding the exact calculations and they have to be sought from the country’s tax legislation, in which case a trip to the tax department is most definitely called for) using the following steps: 1. We take the value of what we’ve received. 2. We subtract, in the form of deductions, what it has cost us to create that value. 3. We multiply what’s left, which is our taxable income, by a tax rate; and the answer is the current tax that we have to pay. Here, however, the value of what we’ve received is what the sum of the cost oil and profit oil that we’re entitled to is worth. So, it’s whatever we sell that oil for on the open market. Then, deductions relate to all the money we’ve spent—remember that the Contractor finances the program of work; so all the money the Contractor spends is deducted (usually, but not always, using the same rules that govern how those same costs are added to the cost pool). And often the few other costs I mentioned just now that the Contractor may also be required to pay—for example, relating to social and educational programs in the country and bonuses—though usually not cost recoverable, are usually tax deductible. Then, from there on, the calculation is exactly the same as before. Sometimes, though, PSCs are written in such a way that the Contractor simply receives profit oil after all taxes have already been accounted for. From the government’s point of view, the thinking is, “If you (the Contractor) get some of the oil and we get the rest, then you might as well only have what you get to keep rather than we give you more than that and then you have to give us some of it back again (in tax).” This is the second part of the earlier sentence for which I issued an advanced warning. In terms of what the Contractor actually receives through the PSC, the end result is, of course, the same thing. Whether the Contractor receives his entitlement of oil and has to pay tax on it or receives only the amount excluding what he would have paid in tax, it’s the same. Production sharing contracts that adopt this approach nevertheless usually explain how to calculate the tax. But why bother if Contractors don’t pay it? There are two answers: one that’s somewhat cosmetic (at least as far as petroleum economists are concerned) and one that means the tax is still important to calculate, never mind the fact that someone else actually pays it. The first answer relates to financial reporting. When oil companies publicly report their oil production and reserves figures, the figures are presented before considering

the effects of tax. In a concession world, they would simply report the oil they’ve produced in the reporting period for their production and what they have in the ground still to produce for their reserves (less, possibly, royalty in some circumstances). In a PSC world, oil companies don’t report that and instead report their entitlement to production and reserves, which is, respectively, the cost oil and profit oil they received in the reporting period and the cost oil and profit oil they still have left to receive over the oil field’s remaining life. Both production and reserves are expressed in equivalent amounts of oil by using actual and forecast oil prices to convert sums of money into equivalent amounts of oil. But the entitlement the companies report is still calculated before considering the effects of tax. So, in this alternative scenario in which Contractors receive their entitlements after all taxes have already been accounted for, they have to calculate what those effects are and add them to their cost-oil and profit-oil entitlements to create an equivalent before-tax version of them. The second answer concerns international oil companies that move money all around the world. Even though they haven’t actually paid it, oil companies have still officially incurred a tax liability. When they move their money—the profits they’ve generated in Country A, for example—from Country A to Country B, the tax these profits have already been subjected to in Country A may be taken into account in determining what tax they will be liable to pay in Country B. So although they don’t actually pay it, that tax may nevertheless have repercussions elsewhere. I’ll say more about this in the third part of this chapter. These tax calculations may be the same as just described (that is, the same as for profit-based taxes in concession fiscal terms), or they may be more straightforward and involve a simple grossing-up of the profit oil (that is taken to be after tax has been paid) into what it would equate to before the tax is paid, assuming a certain tax rate. In Appendix 4.2, I’ve provided some worked examples of these PSC mechanisms (as I have for concession terms in Appendix 4.1), one of which shows the arithmetic behind this simpler grossing-up approach. And that’s PSCs. Service Contracts. As promised, now a few words on service contracts. I said at the beginning of this section that service contracts have much in common with PSCs from a petroleum economist’s point of view, and they do. The difference between them has more to do with the language they use than what they say. Whereas PSCs talk about sharing and profit (as in production sharing and profit sharing), service contracts talk about service and fees (the service that will be provided by a Contractor and the fees that will be paid to him). But in both types of contracts, it’s the oil companies that spend the necessary money and largely execute the agreed work program, and both types have some kind of cost recovery and remuneration mechanisms through which oil companies are compensated. So if you understand in principle how PSCs work, then you understand how service contracts do too.

My closing remark before we move on to the second part of this chapter is this: if the fiscal terms that are relevant to your project are the concession type, then my advice to you is to shut out the contractual type from your thoughts; and vice versa. Thinking about these very different approaches requires very different mindsets, and so it’s easy to get confused. Therefore, whichever applies to you at the time, that’s the world you’re living in at that moment. And that’s all I’m going to say about fiscal terms in the country in which the oil field is located. I’ll move on now to what happens regarding tax if oil companies want to send the money they’ve made in that country to another one. Once again, so far so good.

4.2 Tax on Sending Money to Another Country Our discussion of fiscal terms so far has concerned only those of the country in which an oil field is situated, which in this section I’ll call the “resident” country. But the tax authorities’ reach extends far beyond there. If an oil company is headquartered in, say, the US and owns an oil field in, say, Australia, then it will want to send the profits it makes in Australia back to the US to its headquarters, its parent company. In corporate financial speak, it will want to repatriate its profits. There may be further tax costs associated with doing that, however. Let me go back a step. When a company establishes a presence in a different country from the one in which it’s headquartered, it can do so in one of two ways: it can create a subsidiary there or it can create a branch. A subsidiary is a company in its own right and is owned by another company; but while owned by another company, is separate from it. We say that it’s a separate legal entity. A branch, however, isn’t. A branch is simply another office (a branch office) that the company has in another location. Briefly—because I’m veering slightly off the point of this chapter—a company may be compelled by the country it’s moving into to establish either a subsidiary or a branch there; it may have no choice. Where it has a choice, though, a subsidiary has certain advantages over a branch: • It may demonstrate a greater commitment to the new country. • It has greater independence and can, for example, use local financial markets to raise money. • It can offer more legal protection to the parent company, given its separate legal standing. However, branches are usually simpler to create and administer. Subsidiaries and branches of companies are taxed differently. Both pay taxes in the resident country (reminder: the country in which the oil field is situated) on their operations there in exactly the same way, but when it comes to repatriating profits (reminder: sending profits home to their headquarters), subsidiaries pay withholding taxes and branches pay branch remittance taxes to the resident country’s tax

authorities. These taxes amount to the last throw of the dice by governments to keep companies’ profits in their country before they’re exported, although they work differently from each other. 4.2.1 Withholding Taxes. Withholding taxes are levied on any means a subsidiary might use to send profits out of a country, which include dividends (that is, companies’ payments to shareholders, which here means the parent company, the box in the corporate structure that owns the subsidiary), interest (paid to another part of the company if it has borrowed money from there), royalty (the same principle as the royalty discussed earlier, paid to another part of the company), or payments for services provided by another part of the company. All these payments are, in practice, repatriating profits by another name. Withholding taxes are so called because they aren’t paid by the subsidiary’s owner that receives the, say, dividend income. The subsidiary pays the tax on behalf of its owner in its own resident country, and the latter receives the income after tax on it has already been paid. The tax has been “withheld” in the country in which the subsidiary is resident. It’s like a bank paying you interest on your savings after having already deducted the tax on it, so the interest you receive is less any tax that was due. 4.2.2 Branch Remittance Taxes. Branch remittance taxes are levied on the branch’s taxable income—exactly the same taxable income that was talked about earlier in the chapter—whether or not profits are repatriated. So they’re simpler to calculate because they amount, effectively, to just using a higher tax rate. 4.2.3 Tax Treaties. Withholding taxes and branch remittance taxes help give rise to the kind of corporate structures such as the one shown in the Russia example in Chapter 1 because of tax treaties that exist between countries. Tax treaties mean that these taxes, and others that apply when companies move profits between countries (the next section of this chapter will talk some more about that), can be managed. Tax treaties, which are also called double tax treaties (abbreviated to DTTs) or double tax avoidance agreements, exist between countries or groups of countries to avoid exported profit being taxed twice: once in the country where it was generated and then again in the country to which it’s sent. Tax treaties also therefore serve to encourage trade between the countries involved. In practice, tax treaties minimize and can even eliminate withholding taxes and branch remittance taxes. Here’s that example again (disguised as Fig. 4.1) from Chapter 1: Here, the Russian company is a subsidiary of a holding company in Cyprus (Russia Holdings Ltd.), and loans into Russia are channeled through Luxembourg (Russia Capital Corporation Sarl). A holding company is a company that doesn’t produce anything or provide any services itself but just owns and manages shares in

other companies. These countries—Cyprus and Luxembourg—aren’t just plucked out of thin air. As a result of tax treaties, a Russian company paying dividends to a Cypriot company enjoys reduced withholding taxes on dividends from 15 to 5%, and a Russian company that pays interest to a company based in Luxembourg pays no withholding taxes (on interest) instead of 20%. So that’s why, at least in part, the company is structured like that. The preceding few paragraphs, by the way, are no doubt to tax experts a shocking simplification of their world, but I hope that any of them reading this will forgive me. Of course the subject of tax treaties is much more complicated than I’ve portrayed it here. For example, it isn’t possible, as would seem the obvious strategy, to treaty shop—that is, to simply pick corporate structures that take advantage of the most beneficial treaties around the world—because tax authorities don’t allow it and demand that structures can be justified with sound business reasons.

Fig. 4.1—Example of a company ownership structure and internal cash flow.

To petroleum economists, withholding taxes and branch remittance taxes are relevant cash flows because both are brought about by the new thing the company

does and they cross the border from inside the company to outside it (they’re paid by the company to the resident country’s tax authorities). But, in the Russia example anyway, the cash flows that give rise to withholding taxes aren’t. Dividends, loans and interest on loans, and the repayment of those loans aren’t relevant cash flows. Although they’re brought about by the new thing the company does, they don’t cross over the border; they’re “from one box in the corporate structure to another one.” In practice, withholding taxes can be quite complicated to calculate. Accountingrelated rules govern the size and timing of dividends that companies can pay, and there are rules that govern the size of loans that can be put in place and the interest rates that apply to them. There can also be interactions between these various means a company might use to repatriate profits that have repercussions for petroleum economists (and give them headaches), but these are, I think, beyond this book’s remit and so we’ll leave that discussion for another day. Branch remittance taxes are less complicated, however. As already mentioned, they effectively amount to using an additional tax rate over and above that which is used to calculate resident country tax. And now I hope, once again, that we‘ve reached that point where I’ve talked enough about taxes relating to sending money out of one country to another one such that you now have enough confidence to continue the conversation with your own tax department. We’ll move on now to the third part of this chapter. You’re doing well, by the way. You’re nearly there.

4.3 Tax to Pay in the Other Country Here’s the situation: we’ve been a combination of clever and a little bit lucky with our exploration program in Russia and have made a discovery, which we’ve developed and is producing oil. We’re now making some money and are feeling very pleased with ourselves. However, let’s now say that our Russian company that owns the oil field is a subsidiary whose ultimate owner is an American parent, and we need to repatriate our Russian profits back to the US. We’ve been responsible corporate citizens in Russia and paid all our taxes there. We’ve also paid withholding taxes in Russia to take our profits out of the country, largely in the form of dividends from our Russian company to a holding company in Cyprus. A tax treaty between Russia and Cyprus has reduced those withholding taxes relative to what they would otherwise have been. So some profits in the form of dividends have arrived in Cyprus. Is there Cypriot tax to pay on those dividends? We pay another visit to the tax department, whose experts answer the question as follows: “Every country in the world applies some system of double tax relief, which recognizes that companies exporting profits from one country to another have already paid taxes in the country where the profits were generated and shelters them from being taxed again.” They go on to explain, “There are two types of regime that govern how income generated in one country is taxed when it’s received in another one: the

exemption regime and the foreign tax credit regime. Nowadays, most countries use the former and don’t tax the receipt of dividends from overseas subsidiaries and branches; the income is exempt and no further taxes are paid. Cyprus is one such country; so no, there’s no Cypriot tax to pay.” Then helpfully, despite having already answered the question, they go on to explain further: “One or two countries still use the latter, the foreign tax credit regime. In those countries, tax is paid on overseas income, but the tax already paid overseas is recognised and subtracted from it; the overseas tax is credited against the tax to be paid in the country receiving the income. This regime can therefore even eliminate further tax where tax rates in the profit-exporting country are the same as or higher than in the profit-importing one, although doing so doesn’t, unfortunately, give rise to a tax rebate where the profit-exporting country’s tax rate is higher.” Once again, we leave the tax department with the satisfaction of knowing more than we did when we entered it (even if we didn’t completely understand everything they said). I’ve said it once and I’ll say it again, and I make no apology in doing so: if this situation of sending money from one country to another applies to you, go and see your tax department. And when you get there, look the experts in the eye and ask them with confidence and conviction, “Are there any tax costs of repatriating our profits?” If the answer is no, then you don’t need to care why not (unless you want to, of course). And if the answer is yes, then you have a relevant cash flow on your hands (and possibly more than one of them) and, believe me, you’ll make your tax department’s day by asking them to explain to you where it comes from.

4.4 Indirect Taxes All the taxes we’ve discussed so far are direct taxes, which are taxes on the income we earn and we pay to the government, where “we” can be individuals or whole companies. Income tax is the obvious example that people pay based on their salaries and oil companies (and indeed other companies) pay on the basis of their taxable income. Indirect taxes differ from direct taxes in two ways: • First, they aren’t based on the income we earn. Instead, they’re based on what we buy and sell; which is why they’re often called “transaction taxes” because they’re tax on a transaction—that is, someone selling something and someone else buying it. • Second, the individual or company that pays the tax isn’t necessarily the same individual or company the government collects the tax from. This is most easily explained with an example. Imagine a geologist buying a pint of beer in a pub (I know that’s a hard thing to imagine, but try), and we’ll say that this rare event takes place in the UK. When the geologist buys the pint of beer, part of the price he pays for it is tax. The geologist pays it, but it’s collected by the pub, which gives it to the government.

The same thing happens in oil companies. If value added tax (VAT, a common indirect tax around the world) applies to a barrel of oil, the effect it has is to increase the barrel’s price. Then, when the barrel is sold, the oil company gives the government the amount that represents VAT out of what it receives in payment and keeps the rest, which leaves the oil company no better and no worse off and the buyer of the barrel having paid the tax. Whether a petroleum economist cares about indirect taxes depends on whether his oil company is—to return to the example of the geologist buying a pint of beer—in the position of the pub or the geologist. If it’s in the position of the pub and an indirect tax simply flows through it (“flow-through” is in fact how the event is technically described), then petroleum economists ignore it (although both the tax the buyer pays the oil company and the (same) tax the oil company gives the government are relevant cash flows, they cancel each other out and so are ignored in practice). If the company is in the position of the geologist, however, indirect tax represents an actual cost it must pay and therefore petroleum economists must include it. Value added tax is just one of very many indirect taxes around the world. Some (and only some) other examples are sales and use tax, excise tax, goods and services tax (more commonly called GST), stamp tax, and customs duties; and in fact withholding taxes are generally thought of as indirect rather than direct tax. But how do petroleum economists know what indirect taxes apply in a country and they need to take into account? Yep, you’ve guessed it: by going to see their tax department. So it’s another chance to look your tax experts in the eye and speak with confidence and conviction, this time to ask, “Are there any indirect taxes that we will own the liability for?”

4.5 To Summarize This has been a tough chapter during which you’ve probably drunk a lot of cups of tea, and I apologise for that. But it has been an important one because tax is an enormously important cash flow in petroleum economics. It’s also an enormously complex subject that I’ve no more than scratched the surface of here. I hope, though, that I’ve said enough to outline what associated cash flows should be looked out for, but not too much to confuse or bore you. Governments earn income from oil fields and oil companies in a number of ways. In the country in which an oil field resides, there are three kinds of fiscal terms the country uses to share, along with oil companies, in the income an oil field generates: concession and two types of contractual arrangements (production sharing contracts and technical service contracts). In concession fiscal terms, oil companies are taxed on the difference between the value of the oil that the oil field produces and they sell and the amounts they can deduct associated with the costs they incur. They may also have to pay the government a royalty and other taxes that are production rather than profit based. The key questions relating to the nature of concession fiscal terms in a given

country are: • Is there a royalty or any production taxes? • If so, how do they work? And for the profit-based tax component(s) in particular, additional questions are: • What costs are expensed and depreciated? • For costs that are depreciated, how are they depreciated and when can depreciation start? • For how long can we carry forward taxable losses? • Can we carry back taxable losses? • What ring-fence applies? If it isn’t only around the subject oil field, what other taxable income (or loss) is being generated inside it with which this oil field’s can be consolidated? • What is the tax rate? In contractual fiscal terms, oil companies are a party (called “the Contractor”) to a contract with a country’s government (the other party), which describes certain activities that are aimed at establishing, or raising, oil production inside the area it applies to. The Contractor pays all the costs associated with those activities and is remunerated through cost oil, which pays him back the costs he’s incurred, and profit oil, which is a share of the profits that have been made. The Contractor may also have to pay tax on the profit he makes and, often, also pay other costs that the contract specifies. The contracts explain how cost oil and profit oil are calculated and sometimes, though not always, also explain what taxes, if any, are payable by the Contractor. When an international oil company makes profits abroad from the country in which it’s headquartered, repatriating the profits often incurs further tax costs. Depending on how its business abroad is set up, withholding taxes or branch remittance taxes may be payable when profits are exported, and income tax may be payable when they’re imported again into the oil company’s home country or another one. However, tax treaties between countries and double tax relief reduce, and can even eliminate, these taxes. Oil companies might also be subjected to indirect taxes such as VAT and customs duties on what they buy and sell, which are many and varied in nature in countries around the world. Petroleum economists ignore these taxes in circumstances in which they simply flow through oil companies; when they don’t, however, petroleum economists must take them into account. Tax is a vast and hugely complex subject that is absolutely and necessarily the responsibility of tax professionals, and so the path to their door should be a very welltrodden one for petroleum economists. Now, whether or not you’ve sought solace in one or more cups of tea during this chapter, you certainly deserve one now.

Chapter 5

Net Cash Flow We now have all the individual relevant cash flow components we need to derive the overall impact that the new thing a company may do would have on the company’s total cash flow. Some cash flow components are based on forecasts of how much oil will be produced and the price it’ll sell for (so, between them, the income the company will receive); some are estimates of what it will cost to explore for the oil field, and then if we find oil, appraise it, develop it, produce oil from it, and then decommission it afterwards; and some result from calculations of what the government (or governments) will earn from our exploits.

5.1 Putting All the Cash Flow Components Together Now we can put all the parts together by adding all the income-related relevant cash flows and subtracting all the cost-related ones. The result is the net cash flow: the net effect of all the individual relevant cash flow components put together. And if we do that for each and every year of a project’s life, we can see the net cash flow profile that the new opportunity generates for the oil company. For the whole life of an oil field, from exploring for it through to abandoning it, a net cash flow profile usually looks like Fig. 5.1:

Fig. 5.1—Example of the net cash flow profile for an oil field from initial exploration to decommissioning.

In the first few years, the company spends money on exploring, appraising the discovery it’s made (assuming it’s made one, of course), and then developing it. At Point A, however, income from the oil field starts to exceed spending on it because

the costs of developing it are tailing off and oil production has begun and is increasing. Income continues increasing as oil production increases, doing so all the way to point B, when it begins to decline again. In a tax/royalty world, that would be because of oil production starting to decline, although in a production sharing contract world it would have more to do with development costs having been recovered by then and cost oil therefore reducing, and so the Contractor’s entitlement revenue— which is, remember, cost oil plus profit oil—starts to fall. The cash flow graph presented in Fig. 5.1 is set in concession fiscal terms, which you can tell because royalty is one of the cash flow components (in a production sharing contract world, royalty is taken by the government “off the top” before the contract’s cost oil and profit oil are calculated, and so it doesn’t feature in the Contractor’s entitlement cash flow). You can also tell because decommissioning costs are benefiting from a tax deduction at the moment they’re incurred (whereas in a production-sharing contract, any compensation that decommissioning earns the Contractor occurs only while the field is producing).

5.2 To Summarize Net cash flow is the result of adding together all the income-related relevant cash flows and subtracting all the cost-related relevant cash flows. Typically in an oil field’s life cycle, net cash flow is negative at first while exploration, appraisal, and development costs are being incurred, then it turns positive as oil production starts up and gathers pace, and then it turns negative again at the end of an oil field’s life as a result of the cost of decommissioning.

Chapter 6

Introducing Discounting We’ve talked a lot about cash flow. Most of this book so far has been about it, in fact. There’s still one cash flow we haven’t so much as mentioned yet, however, which is the cash flow between a company and the organizations and individuals that provide the financial backing to it. This chapter is about just that. But let’s take a step backwards first.

6.1 Sources of Money for a Company A company works by accessing money, investing it in its business, and earning more from doing so than the cost of accessing the money in the first place. The money it accesses comes, in a general sense, from two sources: lenders, which usually means banks, and investors. Banks lend a company money, charge interest on the loan they’ve made to it, and fully expect to get their money back again because a loan is secured against the company’s assets, either explicitly through the loan agreement or implicitly as part of the legal process that ensues in the event the company defaults (if it fails to keep up interest payments or pay back the loan) or, in the worst case, goes bankrupt. So banks usually can’t lose out. Investors, on the other hand, can. They put their own (or other people’s) money into a company in what amounts to a leap of faith that they’ll ever see it again, never mind make a profit from it. If the company fails and vanishes into thin air, their money can vanish into thin air with it.

6.2 Return on Investment Not surprisingly then, the reward, or return, they expect on the money they hand over to a company is different for banks and investors. Banks are content with a smaller prize than investors because they take less risk, given the protection they have against losing their money; whereas investors rightly want to be more richly rewarded for taking the real and sometimes significant risk they face of losing theirs. Return is most commonly expressed as a percentage rate. If you borrow money from a bank to buy a house, the bank will charge you an interest rate, 5% per year, say. Investors in a company don’t charge an interest rate, but they nevertheless expect to make a certain return on the money they invest, which they do through receiving dividends as shareholders of the company—providing the company makes money and pays dividends, of course—and through growth in the value of their investment if the company becomes more successful and therefore worth more, thereby making its shares—or stock, it’s the same thing—worth more.

The majority of companies make use of both sources of money depending on a number of considerations, some of which are more practical in nature, such as how much they need, when they need it, and how long they need it for. If, for example, a company needs a relatively small amount of money for a relatively short time, then borrowing it from a bank makes sense. Other considerations are more profound, however, such as whether the balance the company has struck is a matter of concern for those associated with it. For example, a company with excessive debt financing from a bank is a concern because the obligation to pay it all back is a worrying burden on its finances; but then again, a company that’s mostly financed by investors is a concern as well because that’s an expensive way to raise money. So what return do banks and investors expect to earn? One at a time: banks first. The interest rate that a bank charges a company is known. The bank tells you what it is when you take out the loan, although it may not be fixed for the whole duration of the loan because interest rates go up and down. Investors, however, of which there may be thousands in the case of very large companies, don’t all individually tell the company what return they expect when they invest their money in it. That would be unmanageable to say the least. Instead, it’s assumed that every investor in the company expects the same return. Now, at the risk of making this seem like a proper textbook … in 1970, a nowrenowned economist by the name of William Sharpe wrote a book called Portfolio Theory and Capital Markets, in which he published his capital asset pricing model (CAPM for short), which could be used to derive the return that investors should expect to earn from companies they invest in. The model, which is still much used today, captures three elements that together make up the reward that investors should reasonably expect: • The first is compensation for locking up their cash in an investment that may not immediately produce any income, whereas they could have invested it somewhere safe and at least be compensated for the effects of inflation (in government bonds, for example). • The second recognises that the market is a notoriously risky place to invest, and so investors who take that risk should demand a suitably attractive return that rewards them for doing so. • The third recognizes that the market is a big place containing very many companies, some of which are more risky relative to the market as a whole than others. And here (Fig. 6.1) is that model: So the returns that banks and investors expect to earn are either known (in the case of the former) or can be established (in the case of the latter). It would be helpful, however, if these separate returns could be combined into a single one, because then a company would know what overall return it has to make from the money it invests in its business to, at the very least, pay its lenders and

investors what they expect back. And that can be done with some simple arithmetic.

Fig. 6.1—The William Sharpe (1970) capital asset pricing model.

A company knows how much the bank has lent it, and it knows what interest rate the bank charges. And it also knows how much investors have invested in it—which isn’t the money they actually invested but how much their investment is worth now, which is all the shares that investors own multiplied by the share price now—and it knows what return they expect. Then the combined, single, overall return is the average of these individual returns proportionately weighted by how much the bank has lent and how much investors have invested.

For example, a bank lends USD 10 million to a company that investors have invested USD 40 million in (perhaps there are 1 million shares, each of which, right now, is worth USD 40). Let’s say the bank charges 5% interest and investors expect a return of 15%. So one-fifth of the total (USD 10 million out of USD 50 million) must return 5%, and four-fifths (USD 40 million out of USD 50 million) should return 15%. Then, one-fifth of 5% plus four-fifths of 15% equals 13%. So if the company makes a return of more than 13%, then both the bank and its investors are happy. If it makes between 5% and 13%, the bank will be happy, but its investors won’t be (because the bank will get the return it wants, but its investors won’t); and if it makes less than 5%, then they’re both unhappy for obvious reasons. There may be one adjustment that’s made to this calculation, and that’s that the interest rate the bank charges is expressed after tax if the company (the headquarters of it, that is, the part that does the borrowing on behalf of the whole company) resides in a country whose fiscal terms allow interest to be deducted. That’s because, then, the actual cost to a company of borrowing money from a bank is the cost after tax is taken into account. In that case, if, say, the bank charges 5% interest and the country in which the headquarters of the company resides has a tax rate of 30%, then that 5% interest equals 5% × (1 − 30%) after tax, or 3.5%. So the company actually pays only 3.5% interest because the other 1.5% is saved in tax. Therefore 3.5%, rather than 5%, would feature in the preceding paragraph’s calculation. This applies only to the lender part of the calculation and not the investor part, however, because whereas interest may be deductible against tax, dividends never are. This situation is much more likely to occur in countries with concession rather than contractual fiscal terms, and happens, for example, in the US. Now is probably a good time to introduce some particular language relating to all this. The loan from a lender and the investment by an investor are, to a company in receipt of them, capital, which it invests in its business. (Unfortunately, that’s the same word I used in Chapter 4 about tax to describe a kind of cost—capital costs— which can’t be helped I’m afraid.) The interest rate charged on a loan by a lender and the return an investor expects on an investment are, to a company paying them, its costs of capital. And the lender’s and investor’s combined, single, overall return is, to a company, its weighted average cost of capital, which gets shortened to WACC. These are all important terms that I’ll use from now on. A company’s WACC is a relevant cash flow. Its corporate finances are almost bound to be affected by a new project. It might borrow more and it might, in time, pay out more in dividends as a result of it, for example. But there’s a problem. The problem is that such decisions—whether to borrow money or issue more shares—are usually taken at the level of the whole company and not at the individual project level. So isolating exactly what future cash flows relating to these costs of capital are associated with an individual project is close to impossible. So, because of that, petroleum economists can’t treat their companies’ costs of capital in the same way as they treat other relevant cash flows. But there’s a way around the problem because one thing they do know is that any

investment a company makes has to return at least its WACC. Cue discounting.

6.3 Discounting Implicit in discounting is the pretence that the capital a company’s lenders and investors have provided is allocated to the company’s individual oil fields and other projects. As I said just now, no such thing really happens; it’s just pretending. We don’t know how much is allocated to each one, nor, in fact, do we need to know. Then the pretence continues with those oil fields and other projects paying out a return on the capital they’re allocated. Discounting, in fact, is just a different way of looking at something you’re probably already familiar with if you have a savings account at a bank. You provide capital to a bank, which the bank then invests and pays you a return on the capital you’ve lent it. For example, you lend the bank USD 100 and the bank pays you 5% interest; so after the first year you have your original USD 100 and USD 5 interest, making USD 105. Then after the second year, you have the USD 105 from the end of the first year plus, this time, USD 5.25 (5% of USD 105), which equals USD 110.25 and so on. Discounting is the same thing, just using different actors, and applied in reverse. Look what happens if the above arithmetic is done differently; if instead of starting with what you have now and calculating what you will have next year you start with what you will have next year and calculate what you have now, like this:

USD 105 is what you will have at the end of the first year and 1.05 is 1 plus the interest rate. And after the second year:

USD 110.25 is what you will have at the end of the second year, and there are two 1.05’s now, multiplied together, because the USD 110.25 includes 2 years’ worth of interest. Both times you end up with what you had in the first place and the interest has been stripped out. Now let’s substitute the stand-ins with the main actors again. Say a company forecasts that it will receive a cash flow of USD 10 million from one of its oil fields next year. The pretence is that that oil field is allocated some of the company’s capital that lenders and investors have provided and on which it has to pay out a return. It doesn’t matter how much the oil field is allocated; only the return it has to pay out matters. But what return is it? It’s the company’s WACC of course, because that’s the minimum return companies have to achieve to pay their lenders and investors what they expect back. So dividing the cash flow the company will receive next year by its

WACC (plus 1) strips out what the oil field has to surrender to the company’s lenders and investors.

If the company receives a cash flow of USD 10 million again the year after, then, in the same way, dividing that cash flow by the company’s WACC (plus 1), twice this time, multiplied together, again strips out what it has to surrender to the company’s lenders and investors.

So USD 0.9 million next year and USD 1.7 million in the year after are what the oil field will pay toward the company’s cost of capital in those years, and USD 9.1 million and USD 8.3 million are what it will have left over after having done so. And that’s discounting.

6.4 To Summarize Companies source the capital (money) they need to invest in their businesses from lenders (banks, for example) and investors. But that capital costs them money in the form of interest on loans (paid to lenders) and dividends (paid to investors). How much it costs them is either known (the lender says what it will be in the case of a loan) or can be derived (William Sharpe’s capital asset pricing model produces the return, in theory, that investors should require from their investment). A company can aggregate these individual costs of capital into a single weighted average cost of capital (WACC) by averaging them in proportion to how much is provided by each source. That represents the minimum return the company has to achieve on the capital it accesses and invests to at least pay its lenders and investors what they expect back. Although costs of capital are relevant cash flows, they’re different from the likes of an oil field’s revenue, development costs, and operating costs because they usually aren’t dictated by individual projects. Instead, they’re managed at the level of the whole company, and for that reason they’re handled differently by petroleum economists—by discounting. Discounting strips out of any year’s future net cash flow generated by a project the amount it will have to contribute to the company’s WACC. The calculations produce what’s left over after that amount is stripped out by dividing the net cash flow in any year by the company’s WACC plus 1, multiplied by itself as many times as there are years between now and when the cash flow occurs.

Interlude My objective so far has been to lay the foundations of petroleum economics that we can now build upon. Each chapter from now on focuses on an issue relating to petroleum economics, which we’ll talk about, including how petroleum economists approach it and deal with it. So, to some extent, in the first part of this book we’ve talked about petroleum economics in theory and now we’ll talk about it in practice. The issues we’ll talk about are these: • Cash flows in practice (Chapter 7) • More on discounting (Chapter 8) • Other financial measures (Chapter 9) • Inflation and real and nominal cash flows (Chapter 10) • Foreign exchange rates (Chapter 11) • Building and using discounted cash flow models (Chapter 12) • Incremental economics (Chapter 13) • Exploration (Chapter 14) • Sensitivity analysis (Chapter 15) • Acquisitions and divestments (Chapter 16) • Leasing vs. purchasing (Chapter 17) • Presenting your work (Chapter 18) • Going about your business (Chapter 19) I’ll also offer some advice for non-professional petroleum economists who come to petroleum economics from other disciplines (Chapter 20), and I’ll conclude with a final word (Chapter 21).

Chapter 7

Cash Flows in Practice Chapter 2 (What Is It About Cash Flows?) introduced relevant cash flows, which are cash flows that result from a company doing something new (drilling an exploratory well, for example) and also cross its border with the outside world (meaning that cash transfers from a bank account it owns to someone else’s, or the other way around). Relevant cash flows may be brand new; that is, they don’t already exist, such as the cost of drilling the exploratory well, for example, or they may be changes to already existing cash flows, such as a change in how much tax the company pays. However, as is often the case, the devil is in the detail. So this chapter is going to talk about the devil in the detail of relevant cash flows.

7.1 The Difference Principle Petroleum economists aim to establish the difference between their company’s future net cash flow if it does what it’s considering doing and if it doesn’t do it: if it drills an exploratory well and if it doesn’t; if it develops an oil field and if it doesn’t; if it buys an oil field and if it doesn’t; if it sells an oil field and if it doesn’t. But, to borrow again from Chapter 2, there are too many cash flows in even a small oil company for petroleum economists to perform that subtraction properly. So instead, they establish the difference directly without reference to the two total cash flows that include and exclude the something new. Sometimes, and indeed most of the time, that’s straightforward to do. There are cash flows that would clearly exist if, say, the company developed an oil field and clearly wouldn’t exist if it didn’t develop it: income from the sale of its oil if the field was developed and the cost of its development wells and production facilities are obvious examples. But sometimes it’s less straightforward. What about the cost of the team that executes the project? Is that a relevant cash flow? And does it make any difference if the team is newly hired for this project or if the people making it up already work for the company? Or if the company has another project that the same team might manage even if it doesn’t manage this one? Is the cost still a relevant cash flow then? And what about the cost of the office space the team occupies? Does it make any difference if it’s newly acquired space to accommodate this project or if the company already owns it and will, furthermore, continue to own it whether or not the project goes ahead? And what about the cost of the company’s directors who decide whether to approve the project? The truth is that relevant cash flows can be tricky to determine for reasons that will become apparent as I explain the rule of thumb I use to distinguish them in

circumstances like these. My rule of thumb goes like this: • The company has two choices: to go ahead with the new opportunity (whatever it is) or not go ahead with it. Any other opportunities it may have are irrelevant because they will be evaluated separately in exactly the same way. • Again, irrespective of any other opportunities the company may have, relevant cash flows here relate to the resources (for example, people and office space) the company needs if the opportunity goes ahead but doesn’t need if it doesn’t go ahead. So, if the company doesn’t already have the resources it needs, it will acquire them if the opportunity goes ahead but won’t if it doesn’t. And if it already has the resources it needs, it will keep them if the opportunity goes ahead but release them if it doesn’t. So let’s revisit the examples from just now—and we’ll assume that the people who would make up the project team if the opportunity goes ahead are already employed by the company and that the office space they’d occupy is in a building the company already owns. 7.1.1 The Project Team. The project team would be kept if the opportunity goes ahead but wouldn’t be needed if it doesn’t and so, as far as this project, and this project alone, is concerned, the team would be released. (Of course, if the opportunity doesn’t go ahead but another one might, the same process is exercised in the evaluation of that other opportunity and may result in the team being needed after all.) So the project team is a relevant cash flow. 7.1.2 The Office Space. The office space the project team would occupy is a relevant cash flow as well, providing that it’s possible to release it if the opportunity doesn’t go ahead. It isn’t relevant if it can’t be released—for example, because it’s an integral part of a building the company owns. This rule of thumb assumes that companies maintain only the resources they need. If resources aren’t needed because an opportunity doesn’t go ahead and can be released, it assumes they will be. 7.1.3 The Directors. The directors aren’t a relevant cash flow because whether or not this opportunity goes ahead, the company needs them (so they say). They do much else besides making a decision regarding this particular opportunity (apparently).

7.2 A Concluding Word You can see now, perhaps, why I said that establishing certain relevant cash flows can be tricky. Doing so depends on circumstances and one can’t, therefore, be completely prescriptive. And this is the devil in the detail of relevant cash flows.

A crumb of comfort, however, is that these costs are generally quite small—at least, relative to the costs of the opportunity itself, which may include drilling monstrously deep wells in outrageously deep water and building cathedral-like production facilities; all of which is why, often, relatively little thought is given to these sorts of cash flows. But I’ve sat in project reviews and listened to debates regarding whether office costs should be included in the evaluation of an opportunity or not; and on one occasion it was even suggested that the cost of the company’s entire office, from top to bottom, should be included. In that instance, the cost of the company’s entire office was significant enough that some very wrong numbers would have resulted from the evaluation had that cost been used, and some very wrong decisions would have been made.

7.3 To Summarize Petroleum economists aim to establish the difference between their company’s future net cash flows if the company does what it’s considering doing and if it doesn’t. The difference principle (the rule of thumb I use) is a way to establish that difference and also be consistent about doing it, and it involves thinking about the needs of the opportunity in hand and ignoring other opportunities the company may have. When petroleum economists evaluate a new opportunity, some of the potential cash flows they must consider are “clean,” but some aren’t. Clean cash flows would clearly exist if the opportunity goes ahead but clearly wouldn’t otherwise. But there are also cash flows that aren’t clean because it’s more difficult to decide whether they’re relevant or not. These cash flows usually relate to resources that would support the execution of the opportunity and also corporate-level resources whose contribution to it is fleeting. The “unclean” cash flows that petroleum economists conclude are relevant in their evaluations depend on circumstances, and therefore there isn’t a universal prescription for doing so; however, generally, the costs of resources that are needed directly to support the execution of the opportunity are relevant cash flows and corporate-level resources aren’t.

Chapter 8

More on Discounting In this chapter I’m going to carry on the conversation we started in Chapter 6 (Introducing Discounting). Chapter 6 was about how companies raise the capital they need from lenders and investors to invest in their businesses and how much that capital costs them. It went on to examine how that cost, despite being a relevant cash flow, doesn’t feature in petroleum economists’ evaluations of new opportunities in the same way that other relevant cash flows do, however, because it isn’t dictated by individual projects and is managed at the level of the company as a whole. Instead, that cost features in a procedure called “discounting,” which strips out of a project’s forecast of its net cash flows the contribution it will have to make over its lifetime to the company’s cost of capital. But there’s more to a company’s cash flows than the sum of its projects’ relevant cash flows.

8.1 Corporate Costs and Project Costs The last chapter ended with a focus on one example of a cash flow that exists outside the universe of projects’ relevant cash flows: its corporate-level resources, which include a company’s corporate departments, buildings, and centralized activities. These corporate-type costs usually aren’t considered to be relevant cash flows by petroleum economists when they evaluate a new opportunity because applying the difference principle (from the last chapter) usually leads to their exclusion; however, they’re costs that nevertheless have to be paid, and projects are the only means that companies have to earn the money to pay them. The same is true of some exploration costs. Exploration may be successful and lead to projects that companies can further invest in and make money from, in which case it represents investment that’s really just a part of the cost of establishing a project (along with its development costs). But what if the company doesn’t establish a project? What if its exploration fails? How is it paid for then? The only way is out of the profits that successful projects generate. Even though they don’t qualify as relevant cash flows, these corporate-type costs and exploration costs that don’t yield a project have something in common with a company’s cost of capital, and that’s that they also fall under the jurisdiction of the company as a whole. So it makes sense to treat them in the same way. And that’s exactly what happens. Projects are asked to generate a certain return that not only covers the company’s weighted average cost of capital (WACC, if you recall) but exceeds it; and what the return exceeds the WACC by is used to pay for the costs that are associated with the

company’s corporate-related resources and assets, as well as its unsuccessful exploration, that don’t and can’t pay their own way. The inevitable next question, then, is how much should a company add to its WACC to cover these additional burdens?

8.2 How Much Should a Company Add? The question of how much a company should add to its WACC to cover corporaterelated expenses and costs associated with unsuccessful exploration is very difficult indeed to answer because there are countless unknowns: what will our corporate costs even be in the future, and how much will we spend on unsuccessful exploration, in both cases after all taxes are taken into account? What will our portfolio of projects be, and how will they perform? We just don’t know. But companies do what they reasonably can to produce an estimate. And they usually decide on a per cent or two as being about the right answer. So, in reality, the rate companies use in their discounting calculations is made up of two parts: • Their WACC (weighted average cost of capital), which acts as a surrogate for the relevant cash flows relating to the interest and dividends companies pay lenders and investors in return for the capital they provide, and • An additional return that projects must also generate to pay for corporaterelated resources and assets as well as unsuccessful exploration that, although not relevant cash flows in the context of an individual new opportunity the company is considering undertaking, don’t and can’t pay their own way. These parts, separately or together, go by a number of names: • Cost of capital, it goes without saying, is defined by the first bullet point above. The words “weighted” and “average” in “weighted average cost of capital” are often dropped in conversation out of laziness, however. • Hurdle rate is the sum of both the above bullet points; the total rate inclusive of both parts. • Discount rate is a non-specific term that might mean anything; although generally when it’s used, what’s meant by it is hurdle rate. • Opportunity cost of capital is the return promised by the next-best opportunity the company has to invest in and therefore represents a gauge against which new opportunities can be measured. (I’ve only ever seen this term in textbooks, however. I’ve never known it used in practice, and so in the spirit of not talking about what isn’t commonly used in petroleum economics, I’m not going to say any more about it.) Companies may have their own, different definitions of these terms, and so it’s worth checking your own company’s rulebook just to make sure you’re talking the same language as everyone else there.

So, now, let’s cut to the chase and look at what, exactly, discounting produces and what its results mean.

8.3 What Does Discounting Produce? Just to help, for illustrative purposes, here’s a project; or rather, here’s the net cash flow forecast, by year, in millions of US dollars, of a potential new project: Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

90

170

140

110

80

50

(40)

Net cash flow is negative (in other words, there’s cash flowing out of the company) in Years 1 to 3 because of the project’s development costs, net cash flow is positive (there’s cash flowing into the company) in Years 4 to 9 while the oil field is producing, and net cash flow is negative again in Year 10 because of decommissioning costs. Here, negative cash flow (that goes out of the company) is denoted by brackets, and positive cash flow (that comes into the company) has none. Let’s assume that the company’s discount rate (by which I mean hurdle rate) is 10%. Then the project’s net cash flow forecast, discounted, is: Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

90

170

140

110

80

50

(40)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(145)

(83)

68

116

87

62

41

23

(17)

Total 72 (80)

Just as in the example in Chapter 6, each year’s net cash flow is divided by one plus the discount rate multiplied by itself as many times as there are years between the year in question and now, Year 1. The total in the box is the sum of the discounted cash flows, and it’s a significant enough number to have a name. In fact, it goes by one of two names—“present value” or “net present value”—which are often used interchangeably, although, strictly speaking: • Present value is the sum of discounted cash flows whose underlying cash flows are all either incoming or outgoing; in other words, are all either receipts or costs, and • Net present value is the sum of discounted cash flows whose underlying cash flows are both incoming and outgoing, as happens in the development, production, and decommissioning of oil fields. In other words, it’s the present value of the positive cash flows net of the negative cash flows that produce them.

These terms are abbreviated as PV and NPV, followed by the applicable discount rate expressed either as a subscript—for example PV8 or NPV10—or in brackets— PV(8%) or NPV(10%). Whatever kind of cash flow they apply to, both values represent the final piece of the cash flow jigsaw. They take a project-specific net cash flow (which is the sum in each year of the project’s relevant cash flows) and then strip out all the other cash flows (costs)—some relevant, some not—that projects must share the burden of because they can’t be allocated to individual projects, and then they add them all up. And what we end up with is the total free cash flow that the project generates; that is, what it earns less every single cost that it has to bear. Both values, you’ll notice, also contain the word “present”—present value and net present value—which is because what they represent is the single amount that the future cash flows from which they’re derived is equivalent to currently, now, today. It’s the amount that a company should be indifferent to accepting as, say, a cheque today rather than earning through the project and its future net cash flow. This brings me to what is probably the most important paragraph in this book, and if there was a way I could make it flash on and off on the page, I would; but I can’t, so I’ll settle for italicising it instead. The net present value of a future stream of net cash flows represents what that net cash flow stream is worth now, today. Therefore, it shows if the company that owns it is richer or poorer because of it, and by how much. It means that a project’s net present value tells a company whether it would be wise or foolish to invest in it, because to do so will make it richer or poorer. And that, if you remember, is what I said right at the start of this book petroleum economists are for: a petroleum economist’s job is to determine whether his company should grow richer or poorer by investing its money in drilling an exploratory well, for example, or building an oil pipeline. In short, by establishing the net present values of the projects they evaluate, petroleum economists determine whether their companies should grow richer or poorer. So in the example project above, the company owning it is USD 72 million richer now, today, thanks to the project because its net cash flow adds USD 72 million of value now, today, to the company. In other words, out of the USD 260 million of net cash flow the project will generate, USD 188 million (USD 260 million minus USD 72 million) will go towards paying the company’s cost of capital and for its corporatetype costs and unsuccessful exploration in the future, and USD 72 million will be spare, surplus. That result is in contrast to these next two projects’ net cash flows, the first of which looks like this: Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 5

Year 9

Year 10

(80)

(160)

(100)

70

150

120

90

60

30

(40)

And when discounted by the same 10% discount rate, looks like this: Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 5

Year 9

Year 10

(80)

(160)

(100)

70

150

120

90

60

30

(40)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(145)

(83)

53

102

75

51

31

14

(17)

Total 0 (80)

But this project’s net present value is zero, which means it makes the company owning it neither richer nor poorer; it adds no value to the company, nor loses it any. Although the project generates a positive cash flow of USD 140 million (which is the sum of the cash flows before discounting), once it pays towards the company’s cost of capital and also pays its share of the company’s corporate-related and unsuccessful exploration costs it has to contribute to, there’s nothing left. That doesn’t make it a bad project, however; after all, it’s paid for itself and made its required contribution to the company’s corporate expenses. And the second project looks like this: Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

50

130

100

70

40

10

(40)

Which, when discounted by the same 10% discount rate, the project looks like this: Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

50

130

100

70

40

10

(40)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(145)

(83)

38

89

62

40

21

5

(17)

Total (72) (80)

Here, taking on this project would make the company owning it USD 72 million poorer (remember that the brackets denote a negative value) now, today, because it would lose the company USD 72 million of value. The project still generates a positive cash flow—of USD 20 million—but what it has to pay towards the company’s cost of capital and its corporate-related and unsuccessful exploration costs is that and more —much more, in fact. Lastly on the subject of the job discounting does in substituting for certain cash flows, here’s another way of looking, in essence, at what it does. Imagine that we derive an oil company’s future net cash flows over time and then discount them at the company’s hurdle rate. These two net cash flows—one discounted and one not— might look something like this (Fig. 8.1): In Fig. 8.1, the vertical axis is net cash flow (the units aren’t important), and the

horizontal axis is years. To the left is now and to the right is many years from now. The blue line is the total net cash flow, the sum of the company’s projects’ net cash flow forecasts over time, and the orange line is the blue line discounted (it doesn’t matter at what discount rate). Like many oil companies, this company has several oil fields and projects. Its oil fields are producing, and although it’s spending money on one or two new projects that are in development, the company is in receipt of significant positive net cash flow, shown by the blue line. Over time, however, its producing oil fields—and indeed its newly developed fields—deplete, and the blue line declines. Perhaps exploration will result in new discoveries and therefore developments, perhaps not.

Fig. 8.1—Discounted and undiscounted net cash flows.

The space between the blue and orange lines is the cash flow that discounting strips out of projects’ net cash flows. So that space represents the cash flow that projects surrender to pay the company’s cost of capital and also to pay for its corporate-related and unsuccessful exploration costs, which you can see stays constant over time, more or less, although it narrows toward the end of the oil fields’ lives when, of course, corporate costs and exploration will by then be pared back to a minimum. The orange line, then, is what projects provide as free cash flow for the company, which it can use for whatever purpose it wishes—to support growth, for instance, or to return to shareholders.

8.4 Yet More on Discounting There are a few more matters relating to discounting to mention while we’re here. The first is the use of different discount rates for different projects, despite what I said in Chapter 6; the second is the behaviour of discounted cashflows, including to different discount rates; and the third is how discounting is performed, because there

are a few variants to its arithmetic that can be used. 8.4.1 Using Different Discount Rates for Different Projects. Section 6.3 of Chapter 6 began like this: implicit in discounting is the pretence that the capital a company’s lenders and investors have provided is allocated to the company’s individual oil fields and other projects. As I said just now, no such thing really happens; it’s just pretending. And that’s perfectly true. Some companies, and large ones in particular, however, do use different discount rates for different projects or types of projects. The reason they do, they argue, is because applying a single discount rate to every project assumes that every project carries equal risk, which is the average risk that the company’s projects all put together represents and so is the risk the company as a whole represents. But, they go on to argue, that isn’t the case. “Our projects aren’t equally risky. An easy project in a place we know well can’t be viewed as having the same risk as one that’s technically challenging in a new country for us. So we’re bound to use a higher return objective for a project like that than the former sort.” And so that’s what they do. They use a higher hurdle rate for projects they believe deserve a higher return objective because they’re more risky, for whatever reason, than projects that are straightforward. How companies adjust their corporate-based hurdle rate to reflect the relative riskiness of the different projects in their portfolio isn’t, in my experience, an exact science however; it involves a good degree of subjectivity. It’s a practice that’s adopted by some companies, though, and so you may come across it. But it doesn’t affect anything I’ve said up to now, nor does it affect anything I’ll say from now on. 8.4.2 Behaviour of Discounted Cash Flows. You can see in all the previous discounted cash flow examples that the further out in time a cash flow occurs, the more severely it’s discounted and the lower its present value is relative to its undiscounted value. In the examples, Year 2’s discounted net cash flow is 91% of its undiscounted amount, whereas Year 10’s is 43%. Clearly then, as far as present values are concerned, companies want to ensure that short- to medium-term cash flows are understood and correct with more certainty than medium- to long-term cash flows. The implications of which are that a project’s development costs deserve more attention than, say, its abandonment costs; as does how an oil field’s production ramps up at the start of its life more than how it declines at the end; as does the oil price next year more than in 10 years’ time; as does deducting those development costs against taxable income—or cost recovering them—sooner rather than later. And this is also why utility companies encourage us to pay our bills monthly rather than at the end of each 3-month period, and why “buy now, pay next year” offers for new sofas appeal to so many people. It’s because cash flow sooner rather than later is worth more to us today. As we stand (or sit) right here, right now, cash flow today is worth more to us than cash flow tomorrow, and it’s worth a lot more to us than cash flow in ten years’ time.

Now look at the following net cash flows, undiscounted and discounted. Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

90

170

140

110

80

50

(40)

(160)

(100)

90

170

140

110

80

50

(40)

1.08

1.08 × 1.08

1.083

1.084

1.085

1.086

1.087

1.088

1.089

(148)

(86)

71

125

95

69

47

27

(20)

(160)

(100)

90

170

140

110

80

50

(40)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(145)

(83)

68

116

87

62

41

23

(17)

(160)

(100)

90

170

140

110

80

50

(40)

1.12

1.12 × 1.12

1.123

1.124

1.125

1.126

1.127

1.128

1.129

(143)

(80)

64

108

79

56

36

20

(14)

8% discount rate (80) Total 101 (80) 10% discount rate (80) Total 72 (80) 12% discount rate (80) Total 47 (80)

It’s the same net cash flow from earlier that, discounted using a 10% discount rate, produced a net present value of USD 72 million. But look what happens when it’s discounted at different discount rates: 8% and 12%. The higher the discount rate, the more severely annual cash flows are discounted and the lower their net present value is; which makes complete sense. Companies with high discount rates have, by definition, high costs of capital (because the market perceives them as more risky) and/or high corporate and unsuccessful exploration costs that burden their projects, which means that more of their project’s net cash flow has to be syphoned off to pay for that, leaving less left over as profit in the company’s pocket. So discount rate is important. It matters that it’s right. But it’s also difficult to get it right. You no doubt already have a sense that discount rates aren’t a thing of surgical precision. It isn’t easy to pin down a company’s cost of capital: interest rates go up and down, as do share prices, and the capital asset pricing model is only indicative of investors’ expected returns. Added to that, the additional return that projects must achieve over and above their-company’s WACC to pay corporate-related costs and for unsuccessful exploration is difficult to estimate. And also added to that, some companies’ use of hurdle rates that (try to) reflect

the relative riskiness of different projects introduces even more imprecision. Then add to all of that that the discount rate a company uses needs to be stable; it can’t swing one way and then the other from one month to the next, because it’s used to make decisions with, and those decisions have to be consistent over time. A project mustn’t be approved today that wouldn’t have been approved yesterday, nor would be approved tomorrow, because one of the factors just mentioned changed overnight. So while companies regularly review their discount rate—usually every year—they have to take a long-term view of the rate they use. It may be considered a little low in one year or a little high in another, but it’s only actually changed if something happens that materially affects it and companies believe is here to stay. 8.4.3 There’s a Time and a Place for Discounting. Discounting is always anchored at a certain place and at a certain point in time because the (net) present values of projects in the same company have to be as comparable as possible so that choices and decisions can reliably be made. The certain place might be all the way back to the company’s headquarters or some holding company that acts as a collection point for the company’s branches and subsidiaries. It would be confusing if an oil company with projects in several different countries calculated their net present values in their respective countries and not beyond them. What if some of them are liable to pay withholding taxes when their profits are repatriated but the others aren’t? Straightaway, they aren’t comparable. And a certain point in time is also necessary, firstly for the same sake of comparability between projects, but also because, otherwise, the cash flows included in a single project’s evaluation would need continual adjustment as the days and weeks pass while the evaluation carries on. This doesn’t mean to say that it’s fixed forever and a day, but it’s fixed for the purpose of the evaluation or exercise in hand, which might be the moment when a develop-or-not decision has to be made about a project or at the beginning of the new year for that year’s annual planning cycle. The point in time at which discounting is anchored is the discount date, which, as its name suggests, is a particular day. Notwithstanding the need to fix the discount date for a specific purpose, it’s generally chosen to also offer a degree of convenience. It’s more desirable, for example, to use 1 January (the beginning of the year) than 19 February, or 1 July (mid-year) than 23 August, because calculations are easier if they’re referenced, say, at the start of a tax year or the start of a production sharing contract’s reporting cycle. 8.4.4 The Arithmetic of Discounting. It doesn’t have to be, but forecasting projects’ cash flows and discounting them is usually based on annual figures, as in the examples earlier. Forecasting production and estimating costs quarterly or monthly is, of course, possible, but it’s very difficult to do accurately beyond the short term; it’s difficult enough to do annually in the short term. So all the cash flows in a given year are added together and expressed as a

single, annual amount, even though they almost certainly occur at different times during the year: some earlier, some later, some in the middle, and some regularly throughout it; which means a decision has to be made. Given that the discount date is defined very particularly, when that single amount is deemed to occur influences how the discounting arithmetic works; and there are a few options to choose from: • Beginning-year discounting assumes the single amount occurs at the start of the year, and so if the discount date is at the start of the cash flow’s first year, then the first year’s cash flow isn’t discounted because it occurs right on the discount date itself; then, the second year’s cash flow is discounted by 1 year, and so on. • Mid-year discounting assumes it occurs in the middle of the year, and so if the discount date is the same as before—at the start of the cash flow’s first year— then the first year’s cash flow is discounted by a half-year (the middle of the year back to the start of the year), the second year’s cash flow is discounted by one-and-a-half years, and so on. • End-year discounting assumes it occurs at the end of the year, and so with the same discount date, the first year’s cash flow is discounted by 1 year (the end of the year back to the start of the year), the second year’s by 2 years, and so on. Here’s the same project cash flow from earlier—again, whose net present value is USD 72 million using a 10% discount rate—still discounted at the same discount rate but assuming beginning-year, mid-year, and end-year discounting. Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

90

170

140

110

80

50

(40)

(160)

(100)

90

170

140

110

80

50

(40)

1.10

1.10 × 1.10

1.103

1.104

1.105

1 106

1.107

1.108

1.109

(145)

(83)

68

116

87

62

41

23

(17)

90

170

140

110

80

50

(40)

Beginning-year discounting (80) Total 72 (80)

Mid-year discounting (see below) (80) Total 1.0488 69 (76)

(160)

(100)

1.04883

1.04885

(139)

(79)

64

111

83

59

39

22

(16)

(100)

90

170

140

110

80

50

(40)

1.04887 1.04889 1.048811 1.048813 1.048815 1.048817 1.048819

End-year discounting (80)

(160)

Total 1.10 66

1.10 × 1.10

0 1.103

1.104

1.105

1.106

1.107

1.108

1.109

1.1010

(73)

(132)

(75)

61

106

79

56

37

21

(15)

Mid-year discounting is a little trickier to apply than the other two because we assume that a given year’s cash flow occurs all at once in the middle of the year. So if our discount date is the beginning of the year, our first year’s cash flow is going to be discounted by six months (from mid-year back to the beginning of the year), our second year’s cash flow will be discounted by eighteen months, and so on. So we have to calculate the six-monthly, eighteen-monthly, and so on, discount factors (the denominators in the above calculations) that are equivalent to a 10%, say, annual rate; which are kindly provided by the following expression: {Inverse logarithm of [the logarithm of (1 + the annual discount rate)] ÷ 2} −1. For an annual discount rate of 10%, that works out to 4.88%. We should then find that compounding the two six-monthly discount factors takes us back to our 10% annual discount factor: [(1 + 4.88%) × (1 + 4.88%)] = 1.1—and hey presto, it does. So our six months’ discount factor that’s equivalent to a 10% annual discount factor is 1.0488; our 18 months’ discount factor is (1 + 4.88%)3 (cubed because I’m discounting three six-monthly periods), which is 1.1537; our two-and-a-half years discount factor is (1 + 4.88%)5, which is 1.2691; and so on. It’s a little trickier, but not much. Mid-year and end-year discounting produce increasingly lower net present values compared to beginning-year, as you’d expect because, of course, of the extra discounting they apply. It really doesn’t matter which of these types of discounting companies use, at least as regards making the choices and decisions they have to make, providing, of course, that the same arithmetic is used all the time in every evaluation. Clearly, projects can’t be compared, and choices and decisions made, if different arithmetic is used among them. My opinion, for what it’s worth, is that mid-year discounting makes the most intuitive sense to use because it seems to me more sensible to assume that cash flows averagely occur mid-year rather than all at the beginning or all at the end of the year. But, as I say, it doesn’t matter that much.

8.5 To Summarize There’s more to a company’s cash flows than the sum of its projects’ relevant cash flows. The costs of its corporate departments, buildings, and centralized activities are all cash flows that petroleum economists usually consider irrelevant when they evaluate new opportunities, but they’re nevertheless costs the company has to pay,

and the only place the money to do so can come from is the company’s projects. The same goes for exploration that doesn’t yield a project, that’s unsuccessful; the cost of that also has to come out of projects’ pockets. So as well as paying their own way, projects must also contribute to these cash flows that are company-enabling—that cover the company’s costs of capital, its corporate-type costs, and unsuccessful exploration. Because none of these cash flows can properly be attributed to individual projects, they’re accommodated in a company’s discount rate, which is called, more specifically, a “hurdle rate” when it includes all these additional obligations. Discounting using a company’s hurdle rate strips out of a project’s net cash flow all the company-enabling cash flows it has to bear, and the sum of its discounted cash flows is the project’s present value or net present value, depending on the nature of the cash flow. These values are therefore the total free cash flow that the project generates—that is, what it earns less every single cost it has to bear; they represent the actual value, or wealth, now, today, that the project adds to the company (if positive) or reduces it by (if negative). Discount rates are important. The higher the discount rate, the lower a project’s net present value, because the higher the discount rate, the more of a project’s net cash flow must be sacrificed to pay for the company’s company-enabling costs. But the rates are also, in practice, difficult to establish accurately and keep current because many of the factors that form them not only aren’t that easy to establish themselves but also don’t keep still. Companies can’t change them whenever any of those factors change, however, because they have to remain stable to ensure consistency in the choices and decisions companies make, and they’re only changed if something material happens that it’s believed is here to stay. A further practice to ensure that choices and decisions are reliably made is always to anchor discounting at a certain place and at a certain point in time. The certain place makes sure that different projects capture all the cash flows, including, for example, costs of repatriating profits that may differ between them; and the certain point in time, called the “discount date,” makes sure that (net) present values don’t have to be continually adjusted over the course of an evaluation and are also most relevant to its purpose.

Chapter 9

Other Financial Measures Net present value isn’t the only measure oil companies use to indicate the financial attractiveness of new opportunities. There are others, several others in fact, and I’m going to talk about some of them in this chapter. I say some of them because although they all have merits to a greater or lesser extent, a few are more commonly used than others, which are unit costs, payback, internal rate of return, and the present value index. I’ll describe these four measures first and then summarize their relative advantages and disadvantages later on, and I’ll include net present value in that as well. Don’t be surprised if you come across other measures than these, however. But first, by way of introduction: all measures that indicate the financial attractiveness of new opportunities fall into one of two groups—those that involve discounting and those that don’t. Of those I’m going to talk about, unit costs don’t involve discounting; internal rate of return and the present value index do; and payback can be applied either discounted or undiscounted.

9.1 Unit Costs Unit costs are entirely logical to look at. You know what you’ll sell a barrel of oil for (or rather, the amount your company forecasts it will sell for), and so it must make sense to work out the cost of producing it. Then, between them you have a sense of whether these are high-margin barrels (margin is the difference between the price you sell something for and what it costs to produce it; so high margin means very profitable) or low margin (not very profitable) ones. Unit costs can be dissected into whatever level of detail a company has. At the very least they can show, for example, why an oil field’s profitability is poor: perhaps capital costs per barrel are very high, or perhaps government take per barrel is especially onerous. There may or may not be something that can be done about it, whatever it is, but it’s helpful information nevertheless.

9.2 Payback Payback is one of the simplest measures of the financial attractiveness of an investment, and therein lays its appeal. It’s no more than an estimate of how long it will take from starting investing to getting your money back again—for example, from starting spending money on developing an oil field to earning the same amount back again once it’s producing. The rule is the shorter that time, the better. Companies, and indeed people,

naturally feel more comfortable if the length of time they think they’ll be out-of-pocket for is shorter rather than longer. A variant of this is to calculate discounted payback, which follows exactly the same principle but uses discounted instead of undiscounted cash flows. In that case, payback derives the point at which net present value is no longer negative rather than when a company is no longer out of pocket in a straightforward cash sense.

9.3 Internal Rate of Return Internal rate of return (usually abbreviated to IRR) is the discount rate that produces a net present value equal to zero when used to discount net cash flows. Here are the same example net cash flows used several pages ago—the ones with net present values of USD 72 million, zero and USD (72) million when they’re discounted using a 10% discount rate—along with the discount rates applied to them that produce net present values of zero. Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

90

170

140

110

80

50

(40)

1.16

1.16 × 1.16

1.163

1.164

1.165

1.166

1.167

1.168

1.169

(80)

(137)

(74)

57

93

66

44

28

15

(10)

Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

70

150

120

90

60

30

(40)

1.10

1.10 × 1.10

1.103

1,104

1.105

1.106

1.107

1.108

1,109

(80)

(145)

(83)

53

102

74

51

31

14

(17)

Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

50

130

100

70

40

10

(40)

1.02

1.02×1.02

1.023

1.024

1.025

1.026

1.027

1.028

1.029

(157)

(97)

48

122

92

63

36

9

(34)

Total 0

Total 0

Total 0 (80)

These discount rates—16% at the top, 10% in the middle, and 2% at the bottom— are those net cash flows’ internal rates of return (they aren’t exactly 16, 10, and 2%, by the way, in case you do the calculations yourself; I’ve rounded them to whole percentages). You can no doubt immediately see a relationship between net present value and internal rate of return, which is: • Net cash flows with a positive net present value have an internal rate of return greater than the company’s (hurdle) discount rate.

• Net cash flows with a net present value of zero have an internal rate of return equal to the company’s discount rate. • Net cash flows with a negative net present value have an internal rate of return less than the company’s discount rate. So companies seek investment opportunities that have internal rates of return higher than their discount rate because they’re the projects that will make them richer. There’s a warning flag I need to raise about internal rate of return, however, which is that it can be misleading in certain circumstances. At the very heart of internal rate of return is a reliance on there being a change of sign in the net cash flows in question from negative to positive—so a period of cash flow out of the company followed by a period of cash flow into it, just as shown in the graph in Chapter 5 (Fig. 5.1). If there isn’t a change of sign, if net cash flows are all positive (which is usually the case when an oil field’s development costs are in the past and the oil field is currently producing oil), then there is no internal rate of return because there’s nothing to make a return on. But there might also be trouble if the net cash flows have several changes of sign, which can happen if an oil field is developed in phases; one part of it is developed and produced for a period of time before the next part is developed, and so on. Then, the internal rate of return’s arithmetic may be difficult to solve, and there may even be more than one solution. If that happens—it doesn’t happen often and I don’t recall ever having been particularly troubled by it, but it remains a theoretical possibility— the right answer is usually an obvious choice, it being the only one that makes sense. A more real problem is when the periods of cash flow out of and subsequently into the company merge together, which is a danger when the period of “cash flow out” is short (just months) and the amount going out is relatively small (both often the case for small incremental developments of already producing onshore oil fields, for example) and net cash flows are annual. Depending on how these periods fall in a calendar year, it may be that the cash flows out are largely offset by subsequent cash flows in in the same year, and so it appears that a small investment in one year gives rise to relatively substantial incoming cash flows over the following years. The result is a misleadingly high internal rate of return. So beware, especially if you’re a decision maker.

9.4 Present Value Index The present value index—usually abbreviated to PVI and including the applicable discount rate, just like net present value, so PVI10 or PVI(10%)—also called “profitability index” and “benefit/cost ratio,” measures how productive a dollar that’s invested is at creating value. A dollar invested in different opportunities isn’t equally productive, but, of course, the more productive it is the better. PVI is colloquially described as measuring “bangs per buck” because it’s a ratio of the value (bangs) you get back for every dollar (buck) you put in. It’s calculated by

dividing the net present value of a net cash flow by the capital investment element of it, itself discounted; so net present value divided by discounted capital. It can also be calculated slightly differently: as discounted capital plus net present value all divided by discounted capital, which just results in a different value for the index, increasing it by one. So, for example, if net present value is zero, then zero divided by discounted capital equals zero, whereas zero plus discounted capital all divided by discounted capital equals one. Here are the calculations for, and results of, the present value indices for our three now-familiar net cash flows. In these calculations, I assume, for simplicity’s sake, that the cash flows in the first 3 years are entirely capital investment, and I also assume a discount rate of 10%. Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

90

170

140

110

80

50

(40)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(145)

(83)

68

116

87

62

41

23

(17)

160

100

1.10

1.10 × 1.10

145

83

Total 72 (80) Capital investment 80 Total 308 80

Present value index = 72 / 308 = 0.23; or alternatively = (72 + 308)/308 = 1.23

Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

70

150

120

90

60

30

(40)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(145)

(83)

53

102

75

51

31

14

(17)

160

100

1.10

1.10 × 1.10

145

83

Total 0 (80) Capital investment 80 Total 308 80

Present value index = 0/308 = 0.00; or alternatively = (0 + 308)/308 = 1.00

Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

(80)

(160)

(100)

50

130

100

70

40

10

(40)

Total (72) (80)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(145)

(83)

38

89

62

40

21

5

(17)

160

100

1.10

1.10 × 1.10

145

83

Capital investment 80 Total 308 80

Present value index = (72) / 308 = (0.23); or alternatively = [(72) + 308] / 308 = 0.77

You’ve probably already worked out that interpreting the index goes like this: • An index value of less than zero in the original calculation method I described earlier and less than 1 in the alternative method means that capital invested has negative productivity and value is destroyed, so the opportunity shouldn’t be pursued. • An index value of zero in the original calculation method and 1 in the alternative method means that capital invested isn’t productive at all, either positively or negatively; it neither creates nor destroys value. • An index value of more than zero in the original calculation method and more than 1 in the alternative method means that capital invested is productive and creates value, so the opportunity should be pursued.

9.5 Comparing the Measures All the measures discussed so far—net present value, unit costs, payback, internal rate of return, and present value index—tell the same story. But the measures that involve discounting tell a fuller story than those that don’t; although even amongst them, they tell it with different nuances. The two that don’t involve discounting—unit costs and payback—are both intuitively sensible to obtain, but they provide limited transparency of a project’s financial attractiveness. 9.5.1 The Limitations of Unit Costs and Payback. Unit costs are interesting if they’re expressed as a single value covering a project’s whole life, but they’re less interesting when they’re applied to its individual years. It may be helpful to know that a project, over its whole life, has capital costs of USD 20 per barrel, but for individual years that will range from next to nothing (in production, if all the facilities and equipment are behaving) to infinite (in development, before production begins). That’s why unit costs are rather more at home in the world of accounting, where the likes of capital costs and tax, which are very irregular cash flows, are smoothed out by accountants over an oil field’s life and then become more meaningful. The enduring criticism of payback is that it says nothing about what happens after

that point is reached. One project may reach payback before another, but the other project may, over its life, generate more value. So it could cause the wrong decision to be made if it’s used to choose between two or more opportunities. That goes for both the undiscounted and discounted versions. 9.5.2 Discounted Measures: The Full Story. All the measures that involve discounting conclude that a company should grow richer, poorer, or neither by investing in a particular new opportunity. But even then, there’s still a part of the story that’s missing, which I’ll illustrate with a few examples. Let’s say the annual net cash flow forecast of a particular opportunity has a net present value of USD 20 million, which is surely a lot of money—although perhaps it isn’t if the company has to invest USD 1 billion to achieve it. Or let’s say that the net cash flow forecast of a particular opportunity has an internal rate of return of 20%, which is surely attractive—although perhaps it isn’t all that exciting if the project is so small in scale that the actual amount the company will grow richer by is inconsequential. Both of these scenarios happen in the real world. Liquefied natural gas projects, for example, require vast investment and generate enormous revenues, and their net present values of hundreds of millions of dollars may appear impressive at first sight. Yet their internal rates of return are probably only slightly higher than the discount rates of the companies involved in them, and their present value indices are probably only just above zero or one (for the former and alternative calculation methods respectively). Similarly, working over an onshore oil well—that is, maintaining the well by cleaning or repairing it—so that it can produce a few more barrels of oil per day may be a great investment from an internal rate of return point of view, but how much richer it makes the company may be insignificant. So there are two different themes that concern petroleum economists and, indeed, decision makers. Both relate to the materiality of the prize that might be earned through an investment, but one relates to it in an absolute sense, and the other relative to the investment that’s made to earn it. Both materialities are sought after, but each of the three measures I’ve talked about concerns only one of them. Internal rate of return says a lot about materiality relative to investment but says nothing about it in an absolute sense. The same is true of the present value index. And net present value is all about absolute materiality but says nothing about how that relates to the investment involved in earning it. So more than one of them is needed to paint the whole picture, which is why petroleum economists need to present—and decision makers want to see—measures that represent both kinds of materiality their company might achieve from an investment.

9.6 To Summarize Net present value isn’t the only measure oil companies use to indicate the financial attractiveness of new opportunities. There are several others, some of which, like net present value, involve discounting—including internal rate of return and present value

index—and some of which don’t, like unit costs; and some sit in both camps, such as payback. Those measures involving discounting provide a more reliable and meaningful insight into new opportunities’ financial attractiveness than those that don’t, although, individually, they still don’t tell petroleum economists and decision makers the whole story. Net present value tells of the absolute materiality of the prize an investment should produce, whereas internal rate of return and present value index tell of its materiality relative to the investment that has to be made to earn it. Both materialities are necessary to know however, and therefore measures that represent both kinds are sought by decision makers and need to be supplied by petroleum economists.

Chapter 10

Inflation and Real and Nominal Cash Flows Inflation is as relevant to petroleum economists in their day-to-day work as it is in their weekly shopping. Just about every cash flow an oil company experiences is affected by it: the salaries of people who work in project and operations teams; the cost of materials; the cost of services that oil companies buy; the cost of renting buildings, and heating and maintaining them; and transportation costs are all affected. And not only their costs, but what oil companies sell is affected as well. So it’s important.

10.1 Inflation vs. Escalation Inflation is responsible for a rise in the general level of prices over a period of time. We’re all familiar with it, and in oil companies it makes all the things I’ve just mentioned go up a little bit each year. Put another way: a dollar today is worth more than a dollar tomorrow. The dollar (or pound or euro or ruble, or whatever currency you use) in your pocket will still be a dollar (or pound or euro or ruble) tomorrow, but what you can buy with it today may cost more than that tomorrow. Being a rise in the general level of prices sets inflation apart from something else that’s often taken to be the same thing, although it isn’t, which is escalation. Escalation also relates to changes in prices, but it concerns changes that aren’t general in nature and are specific to particular circumstances. It’s something else that’s best illustrated with a few examples: there might be escalation in the price of steel due, say, to suddenly higher demand for it in the Far East; or escalation in drilling rig rates because oil prices have risen. Effects like these are separate from inflation. They don’t constitute a rise in the general level of prices; they constitute a reaction to particular circumstances that brings about a change in what might be considered the normal price of something; and because of that, they belong in the domain of cost estimating to consider. So I’m not going to dwell on it here, apart from encouraging you whenever you read a gas sales contract that contains an “escalation” clause saying the gas price will “escalate” at, say, 3% per year (the purpose of which is to compensate the gas seller for rises in the general level of prices) to, in your copy, cross out the words “escalation” and “escalate” and replace them with “inflation” and “inflate.”

10.2 Real vs. Nominal Cash Flow

Back to inflation: inflation is also the difference between cash flows that are in real terms and those that are in nominal or money-of-the-day terms (“nominal” and “money-of-the-day” both mean the same thing, so use whichever one you like. I like nominal personally). The difference between them is that: • A cash flow in real terms is what it looks like if inflation is ignored, as if inflation doesn’t exist, and • A cash flow in nominal terms is what it will look like in reality after taking inflation into account—if our forecasts of inflation are right anyway. So why are there these two kinds of cash flow? Surely inflation is a fact of life, so why pretend there’s no such thing? In petroleum economics, the reason is that most cost forecasts begin life in real terms. Costs are estimated as if they’re being incurred today, even though they may be incurred many years from now. Then, they’re adjusted for inflation. Companies usually have their own standard assumption of what inflation will be, although it may vary depending on the cash flows involved and where in the world they take place. If future costs and prices aren’t adjusted for inflation and are based at a given year, then they’re said to be in that year’s real terms; for example, 2016 real terms. If they’re adjusted for inflation, then they’re said to be in nominal (or money-of-the-day) terms. Some companies side-step inflation completely and leave cash flows in real terms; and that’s fine, although some special care is needed then in a few respects. First, caution is needed in dealing with certain cash flows—for example, gas prices that are contractually fixed in nominal terms (in other words, the gas price is exactly the same year after year and isn’t inflated)—because it means prices will deflate (get smaller) in real terms over time, just like the dollar left in your pocket will. And second, if, for example, projected capital expenditure and tax numbers in real terms are used for corporate planning purposes, they won’t be the numbers the company will actually experience when the time comes. Whether cash flows are in real or nominal terms, they can be discounted, but a golden rule applies. Just like cash flows, discount rates come in real and nominal varieties as well, and the golden rule is that cash flows in real terms must be discounted using a discount rate that’s also in real terms and cash flows in nominal terms must be discounted using a discount rate that’s also in nominal terms. It’s essential; otherwise the resulting discounted values will be wrong. This brings us to an interesting question: is a company’s hurdle rate (as described in Chapter 8) in real or nominal terms? In other words, does it take inflation into account or not?

10.3 Does the Hurdle Rate Account for Inflation? A company’s hurdle rate is, of course, made up of two parts: its weighted average cost of capital (WACC) and an additional return that projects must also generate to

pay for the company’s corporate-related resources and assets and its unsuccessful exploration. Let’s think about these one at a time. 10.3.1 Weighted Average Cost of Capital. A company’s WACC is itself made up of two parts: the part relating to lenders and the part relating to investors. Let’s think about lenders, and let’s think about banks in a different context for a moment first. When you put your hard-earned (or it may be ill-gotten; it’s none of my business) savings into a bank’s savings account, the bank agrees to pay you interest. But the interest it agrees to pay is regardless of inflation. In other words, the bank makes no promises about what you’ll be able to buy with the interest you earn. It may be the case that prices have risen faster than your savings have, and so you still can’t buy as much with them as you could have done when you gave the bank your money in the first place. That means the bank pays you interest in nominal terms. Likewise, when an oil company borrows money from a bank, it pays it a nominal rate of interest. Similarly, the part relating to investors is also in nominal terms. It represents the overall return investors can expect and not the return they can expect over and above whatever inflation is. So a company’s WACC is in nominal terms. 10.3.2 Additional Return on Projects. The other part of a company’s hurdle rate reflects the additional return projects must also generate to pay for the company’s corporate-related resources and assets and for its unsuccessful exploration. One might argue that this additional return has at least a hint of being in real terms about it, because the costs that underpin it will inflate over time from when they were measured. One might also argue, however, that given all the uncertainty surrounding it, that’s a moot point. And it’s the latter argument that prevails, that it’s an indicative, average percentage that accommodates a multitude of sins. So the hurdle rate that oil companies calculate is in nominal terms, and that’s the discount rate they use to discount net cash flows that are also in nominal terms. If net cash flows are in real terms and exclude inflation, however, then the company’s nominal hurdle rate has to be converted into an equivalent rate that’s in real terms as well. And that’s done like this: 1 + nominal rate = (1 + real rate) × (1 + inflation rate). So if a company’s nominal hurdle rate is 10% and inflation is assumed to be 2%, then its equivalent real hurdle rate is (1 + 10%) divided by (1 + 2%), minus 1, which is 7.8%. Therefore discounting net cash flows that include 2% inflation using a 10% discount rate should produce the same answer as discounting the equivalent net cash flows that exclude inflation using a 7.8% discount rate; which they do, and here’s the proof: Year 1

Year 2

Year 3

Year 4

Year 5

Year 6

Year 7

Year 8

Year 9

Year 10

Uninflated (80) Total 103 (80)

(160)

(100)

90

170

140

110

80

50

(40)

1.078

1.078 × 1.078

1.0783

1.0784

1.0785

1.0786

1.0787

1.0788

1.0789

(148)

(86)

72

126

96

70

47

27

(20)

(163)

(104)

96

184

155

124

92

59

(48)

1.10

1.10 × 1.10

1.103

1.104

1.105

1.106

1.107

1.108

1.109

(148)

(86)

72

126

96

70

47

27

(20)

Inflated (2% per year) (80) Total 103 (80)

10.4 To Summarize Inflation, like love (as the song goes), is all around us; it’s a part of everyday life, and that goes for oil companies too. It’s responsible for a rise in the general level of prices over a period of time; unlike escalation, which also relates to rises (or falls) in the level of prices, but not in a general sense over a period of time and instead in a cause-and-effect sense as a result of particular circumstances. So escalation is dealt with differently, by our cost-estimating friends. Inflation also represents the difference between cash flows that are in real terms and those that are in nominal, or money-of-the-day, terms. Cash flows in nominal terms include inflation, whereas cash flows in real terms don’t. Discount rates also come in these same two varieties, and in situations in which cash flows are discounted, both cash flows and discount rate must be of the same variety; so cash flows that are in nominal terms must be discounted using a discount rate that’s also in nominal terms, for example. The hurdle rate that an oil company calculates by adding its WACC and the additional return that its projects must generate to pay for corporate-related costs and unsuccessful exploration is in nominal terms. But it can be converted into an equivalent rate in real terms, and that rate can then be used to discount cash flows that are in real terms and uninflated. Cash flows that are inflated and discounted using a hurdle rate in nominal terms and their equivalent, uninflated cash flows that are discounted using an equivalent hurdle rate in real terms, have the same net present values.

Chapter 11

Foreign Exchange Rates The oil industry is amongst the most global of industries in the world. No matter where an oil field is, the materials, components, and equipment that enable it to produce oil may be supplied by numerous different countries, and even what it produces may be sold abroad. So with money being spent in, and also being received from, different countries, petroleum economists have to deal with cash flows that are in many different currencies. It goes without saying, however, that the net cash flows petroleum economists ultimately derive must be in one single currency. Therefore, they must convert currencies from one—or, indeed, several—to another. And that’s done using foreign exchange conversion rates, which are forecasts of, say, how many Canadian dollars will be equivalent to a certain number of euros in future years, or how many Malaysian ringgit will be equivalent to a certain number of Thai baht.

11.1 Foreign Exchange Conversion Rates Usually, it isn’t a petroleum economist’s job to decide what the appropriate conversion rates to use are. That’s most often done centrally within an oil company (most companies, in fact, draw on external views and forecasts), which are promulgated so that the same rates are used throughout the company, and are periodically updated. But it is a petroleum economist’s job to establish what the native currencies are of the relevant cash flows that make up a net cash flow, where “native currency” means the currency that actually changes hands when an oil company buys or sells something. And what those native currencies are usually changes and grow in number over a project’s life. When a project is no more than a twinkle in an oil company’s eye, even before the exploration well is drilled that might make the discovery that creates it, the currencies involved are usually few because cash flows are generally estimated in whatever currencies cost estimators prefer to use. As the project takes shape, however, it becomes clearer where things are being bought from, and then the number of currencies grows as, perhaps, pipelines are bought from the UK, well materials and equipment are bought from Norway, the FPSO (floating production, storage, and offloading vessel, which is an oil-processing facility in the form of a ship) is bought, remarkably, from Switzerland, and control systems are bought, extraordinarily, from Middle Earth. The native currency of operating costs is usually more predictable though, and in large part is the currency of the country in which the oil field resides because many of

those costs are locally incurred. So too are the native currencies of what’s sold from oil fields. Oil and other liquids are almost always sold in US dollars, whereas gas, typically, is sold in the currency of the country to which it’s supplied.

11.2 On a Personal Note Here’s a short story about foreign exchange conversion rates. I received an email from someone working for the same company as me, but in our Norway office, asking me if I thought it seemed reasonable that changing the US dollar to Norwegian krone exchange rate had tripled his project’s net present value, adding several hundred million dollars to it. I said no, I didn’t think it seemed reasonable and that it seemed quite the opposite, in fact. So I asked him to send me his calculations, which he did. And it was immediately obvious what was wrong. Everything—oil and gas prices, capital and operating costs, and so on—was entered in his calculations in Norwegian krone, which were automatically subsequently converted to US dollars for net cash flow and net present value purposes. The company’s oil price was, naturally, forecast in US dollars per barrel, which the modeller had converted himself to Norwegian krone. When his model performed its calculations, it converted krone back to dollars using the same exchange rate he’d used for his own conversion. And everything worked fine; until, one day, the company’s exchange rate changed but the now-krone-denominated oil price was left untouched. So when the model performed its calculations again, it converted that oil price into a different dollar-denominated price. A higher one. A much higher one, in fact. When originally entered into the model, the company’s oil price forecast was USD 100/bbl and the exchange rate was 6 Norwegian krone to 1 US dollar; so the oil price entered in the calculations (manually) was 600 krone/bbl. Then the exchange rate changed to 5 Norwegian krone to 1 US dollar; however, the oil price in the model was left unchanged at 600 krone/bbl. The model then converted that price to US dollars but now arrived at USD 120/bbl—a USD 20/bbl, 20%, (false) increase. Not surprisingly, the project’s US-dollar revenues skyrocketed, and so, of course, did its net present value. Please be careful with foreign exchange rates.

11.3 To Summarize Relevant cash flows are often apples, pears, and oranges in terms of the currencies they employ; however, their native currencies must all be converted into a single one for the purpose of deriving a net cash flow. That conversion is done using foreign exchange conversion rates, which individual companies usually determine centrally for use in all their petroleum economics calculations.

Chapter 12

Building and Using Discounted Cash Flow Models This chapter takes a well-earned break from the science of petroleum economics to focus on the principal tool that petroleum economists use to generate net present values and other results. Like many other disciplines, petroleum economists use “models,” and they talk about “modelling” a project or an oil field. The models they use take two forms (or at least they should): a qualitative one and a quantitative one. Qualitative models use words and pictures to help petroleum economists understand projects in order to establish what relevant cash flows are associated with an oil field, whereas quantitative ones use numbers and calculations to help them establish what those cash flows tell them about the oil field. Both qualitative and quantitative models are important, and I’ll talk about them separately.

12.1 Qualitative Models Qualitative models are worth their weight in gold (although perhaps that isn’t the most appropriate expression to use given that they don’t weigh very much). They encourage petroleum economists to understand exactly what the real thing is that they’re going to build a model of. They largely create themselves out of the answers to the following five questions: • What is it? • Where is it? • What do we sell? • Where do we sell it? • How does it get there? 12.1.1 What Is It? “What is it?” sums up the opportunity, which may be a new project or a potential one, though it needn’t be. An oil field may be happily producing away, but perhaps we want to forecast its net cash flow for corporate planning and budgeting purposes, or perhaps we want to value it to buy or sell it. If it is a project, we might be developing a discovery, in which case what does that development look like? What wells are we going to drill and what production facilities are we going to install there? Or we might be redeveloping an existing, old oil field, in which case what’s there now and what are we going to add and take away?

12.1.2 Where Is It? X marks the spot, and immediately you can see where the oil field (or potential oil field) is in relation to countries, coastlines, mountains, jungles, deepwater and other existing oil fields (or potential ones). (One would hope that not all these are present in the locality of the subject oil field). 12.1.3 What Do We Sell? Do we sell just oil? Or just gas? Or both? And if we sell gas, do we recover any liquids (e.g., ethane, propane, butane) from it first and sell them separately? 12.1.4 Where Do We Sell It? Answering this question not only locates the origin of certain cash flows (revenue) into the company but it also clarifies how many such cash flows there are. For example, gas produced from a single field may be sold in several places at various prices under different sales contracts. Sometimes, however, it isn’t known where the oil and gas will ultimately be sold—if, for instance, the relevant commercial arrangements aren’t in place yet—in which case some assumptions have to be made. 12.1.5 How Does It Get There? This question’s answer helps to identify the cash flows (costs), if any, that are associated with transporting the oil to where it’s sold. There may be a chain of costs through different pipeline systems, or there may be none at all if the field’s oil is transported through a pipeline that’s also owned by the field’s owners or is sold at the field itself. 12.1.6 What Does a Qualitative Model Look Like? A qualitative model might look something like this, Fig. 12.1. You’ll notice first of all that it’s sketched and normally my notes would be handwritten apart from the graphic (notes are typed here so you can read them in this small version). I prefer to create a handwritten and drawn diagram like this on a poster-size piece of paper to hang on the wall next to me, although usually with much more writing on it: for example, questions I have, references to commercial agreements, and names of people I’ve spoken to or need to speak to. Everything is written on it that helps me understand what and where the oil field is, what it sells, where it sells it, and how it gets there. There’s no substitute for understanding the project, and that’s why qualitative models are so helpful. You’ll also notice that there’s no mention of tax, and that’s because those cash flows may travel all around the world and back again, and so they’re better explained with a different kind of sketch, one depicting a corporate structure such as you’ve already seen in earlier chapters. Before I move on, it’s also worth mentioning the oil field that the sketch in Fig. 12.1 portrays, especially in view of our just having talked about foreign exchange rates. So, here’s an oil and gas field in the Norwegian sector of the North Sea that sells nothing it produces to Norway. It sells crude oil and various natural gas liquids in the UK and gas in Germany, in US dollars and euros, respectively; and it pays to use a

third-party (meaning owned by someone else) production facility and export pipelines to shore in Norwegian krone, and it also pays costs in the UK once its oil reaches there in UK pounds. Then add to that the capital and operating costs relating to the wells and production facilities at the field itself that would likely be a mix of all those same currencies with, probably, a few more besides, and you can see how foreign exchange rates keep petroleum economists on their toes.

Fig. 12.1—Qualitative model of an oil field.

12.2 Quantitative Models There are two basic kinds of quantitative model: off-the-shelf and self-built. The former are petroleum economics packages, of which there are several on the market, and the latter are do-it-yourself calculations that are built by petroleum economists, usually in Excel. Both have pros and cons, and which is better at the job depends on circumstances. On the pros side: • The strengths of off-the-shelf models are convenience, standardization, availability of support and maintenance, and provision of training and documentation, whereas • The strengths of self-built models are flexibility, transparency, intimacy with the calculations (especially for the model’s author), and greater freedom to share

them with others. And on the cons side: the weaknesses of each approach is the other’s strengths. Off-the-shelf models can lack flexibility and transparency, their inner workings can be obscure, and they usually require a license to use them, in which case they can be shared only with others who also have a license. And self-built models can be inconvenient, unique, and unsupported in terms of both ongoing maintenance and education in how to use them. In terms of their integrity, in my experience, both sorts of models may or may not be error-free, and whichever is used, it should be rigorously reviewed and tested to ensure, as far as it’s possible to do so, that its calculations are robust. So circumstances that might make an off-the-shelf model better at the job are if a company uses someone other than a petroleum economist to take that role in an evaluation or if its planning and budgeting work is centered around an off-the-shelf product, both of which are often the case. Then, the strengths that come to the fore are the model’s convenience and standardization. And the circumstances that might make a self-built model a better choice are if some special features are needed, perhaps only temporarily, that may be costly to incorporate in an off-the-shelf model and may take too long, which is a situation a company might face in the middle of a negotiation. Or if a model must be shared with government representatives or other companies that are partners in an oil field, which is sometimes necessary. Then, the strengths that come to the fore are a model’s flexibility, transparency and the ability to share it. Whatever the model, however, more important than anything else is that its user has a sound understanding of petroleum economics. A Rolls Royce being driven by someone who can’t drive will still crash.

12.3 On a Personal Note Over the years, I’ve got into certain habits that work well for me. I always build my own model, which I do because the insight I gain from it is invaluable. I can read a production sharing contract and think I understand it, but I translate its words into arithmetic, and I know I do. I can also read a gas sales agreement, but until I’ve turned the words into calculations, I can’t be sure that I get it. But why does that matter? If a model is available off-the-shelf, why not just use that? For me, it’s all about insight. I’m here to play my part in my company making the right decision and to work out how we can create the most value we can, and to do that I have to have the best understanding I can of what governs how value will be created (and destroyed). So I have to understand the cash flow model inside out and upside down. Another, more pragmatic reason is that, believe me, it’s an unpleasant experience to stand in front of your senior management to explain your results and you’re asked, “How come when that [some input parameter] goes up, that [net present value, say] goes down,” and you don’t know the answer. Although I say that partly in jest, believe

me again, it can damage both your results’ and your own credibility if you don’t know. I use Excel, and I build models that are as simple and straightforward as I can. I don’t use macros (that automate tasks and functions in the spreadsheet), and wherever possible I use nothing more exotic than addition, subtraction, multiplication, and division in calculations that flow in a logical sequence from start to finish. All of which especially helps when I have to share my models with someone else or ask for someone’s advice regarding the integrity of the calculations. Not everyone is an Excel expert (I’m certainly not), and so the calculations need to be as simple and clear as possible. I also do my best to make them neat and tidy without odd calculations here, there, and everywhere. And I add notes that explain the logic of calculations if needs be or point to reference clauses of contracts where values, calculations, or assumptions are defined. Conversely, it’s a frustrating day in the office when I have to audit a model that’s shrouded in macros; and even when I get past them, I find commands that I don’t understand and hard-coded values in cells and formulae without any explanation of what they are and where they came from.

12.4 To Summarize There are two kinds of model in a petroleum economist’s toolkit, a qualitative one and a quantitative one. Qualitative models use words and pictures to provide the understanding of what an oil field looks like and how it works, and therefore of the cash flows it generates. Quantitative models, on the other hand, use numbers and calculations to provide the answer to the sixty-four-thousand-dollar question, “Should doing this make us richer or poorer?” or, perhaps, to forecast an oil field’s future net cash flows for planning and budgeting purposes. Qualitative models need be no more than a hand-drawn sketch, and quantitative models can be bought off-the-shelf ready to use or petroleum economists can build their own, with circumstances dictating which approach is most appropriate for the job. But whatever model is used, more important than anything is that its user is a competent petroleum economist. And with that, it’s back to the science of petroleum economics.

Chapter 13

Incremental Economics Strictly speaking, all petroleum economics is incremental in the sense that only relevant cash flows matter; relevant cash flows being cash flows that are created by the company doing something new (and also that cross its border with the outside world; from Chapter 2). So such cash flows are additive to those the company will experience anyway—they’re incremental to them in other words. Incremental economics is something else, however. It’s a means of deriving certain relevant cash flows, and therefore the net cash flow associated with doing something new, when there are particular circumstances that make it impossible to do by considering the something new on its own, in isolation from the rest of the company. The way we’ve approached evaluating the something new so far makes the assumption that we’d get exactly the same result—net present value, say—by • Evaluating it separately, on its own • Evaluating it by adding it to the rest of the whole company and seeing what difference it makes to the whole company’s net cash flow (by subtracting the company’s total net cash flow before the something new is added to it from the total net cash flow after it’s added) And so we do the former, of course, because it’s much easier. But sometimes we don’t get the same result, and, when we don’t, the reason is almost always because of… (an icy chill descends and the lights flicker …) tax (and there’s a deafening clap of thunder and a blinding flash of lightening). For example, a company produces oil and pays tax. It drills an exploration well. Can we assume the company will remain tax paying when it drills the well? Not necessarily, because the tax deduction that’s associated with drilling the well may mean the company has negative taxable income and won’t pay tax that year. Another company has a production sharing contract and is producing oil from the contract’s area. The percentage of profit oil the Contractor receives depends, let’s say, on the level of total oil production from the contract’s entire area. Then the company makes another discovery inside the area. If we evaluate the new discovery on its own, can we assume the Contractor will continue to receive the same profit share? Not necessarily, because production from the new discovery may mean the Contractor’s share of profit oil for the whole contract area changes. These situations and others like them aren’t uncommon; and in such cases we can’t evaluate the something new on its own, it has to be evaluated in the wider context of the company or the production sharing contract as a whole. And that means the evaluation has to be done the long-winded way, by applying the second

(above) bullet point’s approach—by adding the something new to the whole company and seeing what difference it makes to the whole company’s net cash flow. And that’s called “incremental economics.”

13.1 How Does Incremental Economics Work? In principle, the process involved in incremental economics is no different than that used in petroleum economics in general: we derive a net cash flow and discount it. What is different is how we derive the net cash flow to discount. Until now, that’s involved experts in the company estimating and forecasting certain relevant cash flows and others being calculated, all relating specifically to the something new. Then, all of them are thrown into the petroleum economics pot and stirred to produce the net cash flow the company will experience as a result of doing that something new. Now, that has to be done twice: once including all the company’s other cash flows that may be affected by doing the something new along with the cash flows specifically relating to it, and again including the same other cash flows but excluding the something new. Then the difference between them (the former minus the latter) is the actual net cash flow the company will experience as a result of the something new; and that’s the net cash flow we discount.

13.2 How Do We Know When To Apply Incremental Economics? Unfortunately, no automatic buzzer sounds or warning light flashes when incremental economics is necessary to use. It’s down to petroleum economists to realize when it’s necessary to use it. Petroleum economists have to understand the fiscal terms that apply, what its company’s position is with respect to them, and how the something new might effect that position. But there are some rules of thumb: • If it’s a tax/royalty world we’re in and our company is a significant oil producer and isn’t spending much money, say, developing other oil fields and therefore is a significant taxpayer, and then if we make a relatively modest new discovery that we’re considering developing and producing, it’s safe to assume that developing it won’t have any further-reaching effects on our tax position and it can be evaluated separately, on its own. • And if we live in a production sharing contract world, we produce oil from the contract’s area and the contract has no mechanisms that change what the Contractor receives depending, say, on how much oil is produced, it’s safe to assume that a new discovery we’re considering developing and producing won’t have any further-reaching effects and it can be evaluated separately, on its own. If we’re not in situations like these, however, and any new barrels of oil the company

produces might result in the barrels it already produces being “taxed” differently, then incremental economics it is. It’s better to take a little longer with an evaluation and end up confirming that applying incremental economics isn’t necessary than to take a chance that it isn’t and be wrong.

13.3 An Example of Applying Incremental Economics We’ll end with a worked example, although if words speak louder than numbers for you and you’d prefer to skip it, then I’ll see you in Chapter 14. Otherwise, here’s the situation. A company already produces oil in a country and has made another discovery there that it’s considering developing and producing. The country’s fiscal regime is tax/royalty. The discovery is, however, the only successful well from a programme of several expensive exploration wells that have been drilled. As a result of the exploration programme’s significant expense, the company has accumulated substantial losses for tax purposes and so not only doesn’t pay tax at the moment, but doesn’t expect to do so for some years to come. Here’s what the development and production of the discovery separately, on its own, looks like—at least for the first 10 years. 1. Something new separately, on its own Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 A Production, BOPD

6,000 12,000 9,000 6,750 5,063 3,797 2,848 2,136

B Revenue C Capital

219 350

350

438

329

247

185

139

104

78

120

D Capital depreciation for tax

164

164

164

164

164

E Operating costs

36

47

41

37

34

32

30

29

63

391

287

209

151

107

74

49

0

0

0

0

0

0

(13)

0

0

F Pretax cash flow

(350) (350)

G Taxable income calculations: Opening balance Additions

0

0

19

227

123

45

(13)

107

74

49

Taxable income

0

0

19

227

123

45

0

93

74

49

Closing balance

0

0

0

0

0

0

(13)

0

0

0

0

0

6

68

37

14

0

28

22

15

57

323

250

196

151

79

52

34

H Tax I Post-tax cash flow, undiscounted J Post-tax cash flow, discounted K Internal rate of return

(350) (350) 72 13%

A few accompanying comments are:

• Numbers are USD million unless otherwise stated. • Row A: forecast of oil production from the new discovery in barrels of oil per day (BOPD). • Row B: revenue from oil sales assuming USD 100/bbl sales price every year. • I haven’t assumed any inflation (cash flows are therefore in real terms); it isn’t important for the example’s purpose. • Row C: development capital expenditure. • Row D: tax depreciation relating to development capital assuming 5-year, straight-line depreciation (20% per year, therefore), commencing when the acquired assets are placed into service (which I assume is when production begins). • Row E: forecast of total operating costs during production. • Row F: pre-tax cash flow = Row B – Row C – Row E. • Rows G: ○ Opening balance: losses brought forward from the previous year, if any, expressed as a negative amount ○ Additions: Row B – Row D – Row E ○ Taxable income: the sum of the opening balance and additions providing the result isn’t negative; if it’s negative, then zero ○ Closing balance: opening balance + additions – taxable income • Row H: taxable income multiplied by the tax rate, which, let’s say, is 30%. • Row I: Row F – Row H. • Row J: Row I discounted to the start of the project using a 10% discount rate (and end-year discounting). • Row K: internal rate of return based on Row I. So, separately, on its own, for the first 10 years, the project has a respectable USD 72 million of net present value and a 13% internal rate of return. Now here’s what the rest of the company looks like excluding the development and production of the new discovery. 2. The rest of the company Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Production, BOPD

L

9,000 7,200 5,760 4,608 3,686 2,949 2,359 1,887 1,510 1,208

Revenue

329

263

210

168

135

108

86

69

55

44

Capital

59

41

29

32

17

28

12

9

11

6

Capital depreciation for tax from capital expenditure before year 1 24

19

16

11

Capital depreciation for tax from capital expenditure in and after year 1

12

20

26

32

36

29

24

20

15

13

Operating costs

51

48

46

43

42

40

39

38

38

37

Pretax cash flow

218

174

136

93

76

39

35

21

6

1

(99)

(97)

Taxable income calculations: M

Opening balance Additions Taxable income Closing balance Tax

(850) (609) (433) (310) (228) (171) (133) (109) 241

176

123

82

57

38

23

11

2

(6)

0

0

0

0

0

0

0

0

0

0

(99)

(97)

(103)

(609) (433) (310) (228) (171) (133) (109) 0

0

0

0

0

0

0

0

0

0

Post-tax cash flow, undiscounted

218

174

136

93

76

39

35

21

6

1

Post-tax cash flow, discounted

608

Exactly the same accompanying comments apply as before, with just the following additions: • Row L: tax depreciation relating to capital spent before Year 1 relates to capital spent by the company over the 4 years before commencement of the new project. • Row M: the company’s opening balance of its taxable income calculations in Year 1 is an accumulated loss of USD 850 million (for reasons explained earlier). Tip: it isn’t necessary, therefore, to model oil fields and companies that are midlife from their beginnings; models can start at any time in their lives, providing that their fiscal positions at that point include any carried-forward fiscal attributes—that is, amounts for “tax” purposes that are generated by historical costs and that carry forward into the current and future years—such as the above items. So you can see that the company expects to pay no tax for the next 10 years (because the taxable income it generates going forward serves only to erode most, but not all, of its losses). Now here’s the same information again, but this time with the development and production of the new discovery added to the rest of the company, the consolidated position in other words. 3. Something new added on to the rest of the company Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Revenue

329

263

430

607

463

354

271

208

159

122

Capital

409

391

149

32

17

28

12

9

11

6

Capital depreciation for tax from capital expenditure before year 1 24

19

16

11

Capital depreciation for tax from capital expenditure in and after year 1

Operating costs Pretax cash flow

12

20

190

196

200

193

188

20

15

13

51

48

81

90

83

78

74

70

68

66

199

484

363

249

186

128

80

50

0

0

0

0

0

0

(132) (176)

Taxable income calculations: Opening balance Additions Taxable income Closing balance Tax Post-tax cash flow, undiscounted Post-tax cash flow, discounted

(850) (609) (433) (290) 241

176

142

309

181

83

10

118

76

43

0

0

0

19

181

83

10

118

76

43

0

0

0

0

0

0

0

0

6

54

25

3

35

23

13

199

479

309

224

183

93

57

37

(609) (433) (290) 0

0

(132) (176) 705

Here, the revenue line is the sum of the revenue lines from the previous two sets of information, the capital lines are the sum of the capital lines, and likewise the operating costs line. The initial opening balance of the taxable income calculations comes from the company’s accumulated losses; and then the calculations work themselves out as before, resulting in how much tax the company will pay, a new post-tax cash flow, and a new net present value (post-tax cash flow, discounted). Now we can subtract the post-tax cash flow relating to the rest of the company without the new discovery from the post-tax cash flow relating to the rest of the company but assuming the discovery is developed and produced, to derive the new discovery’s incremental economics. 3. minus 2. Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Post-tax cash flow, undiscounted Post-tax cash flow, discounted Internal rate of return

(350) (350)

63

386

233

184

148

71

51

36

97 15%

So developing and producing the new discovery won’t make the company merely a respectable USD 72 million richer; it will, in fact, make it an even more respectable USD 97 million richer; and instead of it generating only a respectable 13% internal rate of return, it in fact generates an even more respectable 15%. But how has that happened? The answer lies in the tax lines. Considered

separately, on its own, development and production of the discovery would result in USD 190 million of tax over 10 years. The rest of the company, without developing and producing the discovery, will pay no tax in that time (because of its accumulated losses). But overlay the development and production of the discovery onto it and the company would pay USD 159 million, which is USD 31 million less tax. Whereas after ten years some accumulated losses still exist in the company without the developed and produced discovery, with it brought into the company, all those losses are rapidly eroded, and so all the benefit they represent—through reduced tax—is captured in that time.

13.4 To Summarize Incremental economics applies when the relevant cash flows that are associated with something new the company might do are impossible to establish by evaluating the something new separately, on its own, without considering some less obvious, indirect effects that it brings about. Usually, these effects stem from the fiscal terms that apply and happen because the introduction of new costs, new production, and new income to the company sometimes means that different fiscal cash flows result from evaluating the something new separately, on its own, and evaluating it by assessing the difference it makes when it’s added to the company’s existing costs, production, and income. Evaluating the something new by means of incremental economics involves establishing two net cash flows: one including all the company’s cash flows that aren’t specific to the something new but might nevertheless be affected by it, and adding the something new’s own cash flows to them; and the other including only the company’s cash flows that aren’t specific to the something new but might nevertheless be affected by it. Then, the first net cash flow minus the second one is the overall difference the something new makes to the whole company’s net cash flow, and that’s what’s discounted for net present value and other purposes. Petroleum economists have to be alert to when it’s necessary to use incremental economics. If in doubt, consider it necessary.

Chapter 14

Exploration This chapter discusses a generalised approach to the evaluation of exploration opportunities. Individual oil companies may have their own processes and terminology, however. Oil companies explore for oil, and sometimes they find some and often they don’t. And sometimes they find some, but it isn’t enough oil to justify the usually significant cost of producing it, and so they leave it where it is. So how do oil companies decide whether to drill an exploration well when they don’t know what the outcome will be? The answer is by using their professional judgment along with some arithmetic. Petroleum economists can evaluate how much poorer their company will be if it spends money on an exploration well that doesn’t find any oil, and they can evaluate how much richer it should become if it does find some, assuming it is richer, that is. And the company knows what at least it thinks the chances are of each outcome happening. All of which can be combined in some arithmetic to indicate whether the company will, by rights, end up richer or poorer by drilling the well. I’ll talk about the arithmetic later on, but first let’s talk some more about the information that goes into it.

14.1 If the Company Doesn’t Find Any Oil In the event that no oil is found, the company has incurred a cost and has nothing to show for it. But it may be possible to soften the blow of that disappointment, financially at least. It depends on the applicable fiscal terms (you think you hear a scream from somewhere nearby, or was it just the kettle’s whistle?) and the company’s situation with respect to them. In a concession (tax/royalty) world, the cost of the exploration well can be deducted for tax. In reality, however, whether that softens the blow or not depends on whether the company is paying, or is likely to pay, any tax. The well’s cost is usually immediately deductible; so if the company drilling it is paying tax, then the tax it pays is immediately reduced. How much it reduces by depends on the cost of the well and the tax rate; the higher either of them is, the greater the reduction. If the company doesn’t currently pay tax but will in the future, then the exploration well will reduce the tax it pays when that time comes, which discounting renders less impactful but is still beneficial. If, however, the company will never pay tax—because it will never make a profit for tax purposes—then the deduction will never materialize and the blow will never be softened.

In a contractual world, the cost of the exploration well is recoverable. Whether it’s actually recovered, however, depends on oil being produced from the area governed by the contract and, on occasions, even from the individual field that the well discovers (because cost pools are sometimes, although rarely, specific to individual fields). Of course, if no oil is produced, then there’s no cost recovery and, once again, no softening of the blow whatsoever.

14.2 If the Company Does Find Oil In the event the company is lucky and finds some oil, then we—“we” being petroleum economists—can determine how much richer or poorer the discovery should make us —“us” being the company. So how much oil will be produced and sold, the price it will sell for, the costs of developing and producing it, and any other relevant cash flows all have to be estimated and forecasted by experts to provide the petroleum economist with the data his calculations need so that he can establish the field’s net present value.

14.3 Our Judgment of the Outcome “Our” judgment of the outcome isn’t the petroleum economist’s however; this isn’t petroleum economists’ territory. It’s geoscientists’ territory. (Geoscientists tend to think of the local pub as their territory, but that’s in a different context.) Geoscientists provide the probability, from a geological point of view, that oil is present at the well’s target location. So, for example, we may drill an exploration well that our geoscientists deem has a probability of failing—that is, not finding any oil—of Pfailure equals 75%, and has a probability, Psuccess equals 25%, of finding, say, 80 million barrels of oil (the probabilities must add up to 100% of course); as shown in Fig. 14.1.

14.4 The Arithmetic So we know how much poorer we’ll be if the well doesn’t find any oil—and the probability, we think, of that happening; and we know how much richer (or poorer) we’ll be if it does find some—and the probability, we think, of that happening. Then, the overall, expected outcome of drilling the well is the sum of the individual possible outcomes weighted by the probability that each will occur: [value lost if we don’t find oil × Pfailure] + [value gained (or lost) if we do × Psuccess] Returning to our example, let’s say that we’ll lose USD 18 million of present value if we drill the well and don’t find oil and that we’ll gain USD 50 million of net present value if we drill it and find 80 million barrels; so we have a 75% chance of losing USD 18 million and a 25% chance of gaining USD 50 million, in which case the overall, expected outcome is:

[75% × (negative 18)] + [25% × 50], which equals negative USD 1.0 million That means if the well was drilled a thousand times, we would, on average, be USD 1 million poorer. So, we can say that that’s the expected outcome of drilling the well, and, in that case, we shouldn’t drill it because, by the law of averages, doing so will make us poorer.

Fig. 14.1—Example of geoscientists’ estimation of the probability of finding oil.

If, on the other hand, we thought an 80-million-barrel discovery should make us USD 60 million richer, then the same calculation would equal positive USD 1.5 million. On average, therefore, drilling the well will make us richer, and so we should go ahead and drill it. This expected outcome of drilling the well is called the well’s “expected monetary value” and is usually abbreviated to EMV, followed, like net present value, by the discount rate that was used to derive the values of the possible outcomes; so, for example, EMV10 or EMV(10%). In general, relative to the alternative of not drilling the well, which of course has a do-nothing value of zero: • If an exploration well has a positive expected monetary value, drilling it will probably make us richer, and so we should proceed, and • If it has a negative expected monetary value, drilling it will probably make us poorer, and therefore we shouldn’t.

14.5 More Realistically Though … The approach just described may be satisfactory when little information exists about an exploration target, but as more and better information is assembled about it, more

sophistication can be applied to its evaluation. Staying with the same example, let’s say that after further work we’re able to express our inevitable uncertainty about what we might find rather than take it as certain, as shown in Fig. 14.2. Now: • Pfailure (the probability from a geological point of view of not finding oil) is still deemed to be 75%, however • The potential oil field is still thought to most likely contain 80 million barrels of oil, but we recognize now that it could contain as little as 30 million barrels and as much as 200 million barrels.

Fig. 14.2—Geologists’ revised probability of finding oil.

Of course, there are a great many potential outcomes between 30 million and 200 million barrels, but we certainly wouldn’t evaluate every one of them because doing so would be impractical. We can, however, and typically do, evaluate three of them: a low-case estimate, a mid-case, and a high-case. Then, if probabilities that they will occur can be assigned to them, we can adopt the same process and use the same arithmetic as before to calculate the well’s expected monetary value, which admittedly involves more work because there are three success cases now rather than one, but the result will be more informative. The three points that are usually targeted are: • In the low case: the point at which there’s only a 10% chance of a lower outcome, • In the mid-case: the point at which there’s a 50% chance it will be higher or lower, and • In the high case: the point at which there’s only a 10% chance of a higher outcome.

In statistical parlance, these are the 90th, 50th, and 10th percentiles, or P90, P50 (or median), and P10. In our ongoing example, let’s say that these three points are 55 million barrels for the P90, 84 million barrels for the P50, and 125 million barrels for the P10 (the most likely outcome isn’t the same as the P50 here because of the shape of the distribution of possible successful outcomes), as shown in Fig. 14.3. Having reached this point, these three success scenarios next take a test that involves fast-forwarding to when the company must consider whether it would develop and produce each scenario if it finds them, which may be a week or two, a month or two, a year or two, or more away. The baton passes back to the petroleum economist from the geoscientists, who, again with the help of engineers and other experts evaluates the three scenarios to establish whether developing and producing them should make the company richer or poorer. To do that, the petroleum economist anchors these evaluations, not to when the decision to drill the exploration well is taken, but to when the decision to develop the (assumed) discovery is taken—a week or two, a month or two, a year or two, or more afterwards—because, for this test, that’s the decision in hand.

Fig. 14.3—Three success scenarios.

Let’s say in our example she concludes that developing and producing a 55-millionbarrel discovery should make the company USD 23 million poorer (in other words, its net present value is negative USD 23 million), an 84-million-barrel discovery should make it USD 79 million richer, and a 125-million-barrel discovery should make it richer by USD 174 million. But we wouldn’t develop a discovery that makes us poorer, so if we discover the example’s P90 scenario, there will be no development.

14.6 Geological vs. Commercial Likelihood of Success Geoscientists provide the probability, from a geological point of view, that oil is present at the well’s target location. But that oil may not exist in sufficient quantity to persuade us to spend millions of dollars of our money—or, rather, our lenders’ and investors’ money—developing and producing it, as we’ve just seen. So, in fact, there are two measures of an exploration well’s likely success: a geologically based one and a commercially based one. The former represents the likelihood of finding any oil at all, and the latter the likelihood of not only finding oil, but finding enough that would make us willing to develop and produce it. These are called “POSg,” for probability of success, geological, and “POSc,” for probability of success, commercial. Although the former is what’s used in expected monetary value calculations, the latter is often a helpful indictor of an exploration target’s real chances of ultimately making us richer. To approximate POSc, we have to have a few data points on a curve of field size against its associated values, which we already have for our perennial example (see Fig. 14.4). Some fiscal regimes don’t produce a conveniently linear graph like this and instead result in discontinuities where, for example, a production sharing contract’s profit share changes suddenly, in which case another point or two on the graph may be needed. In our example, these values are still forward in time when the decision to develop a discovery or not will be made because to develop it or not is the decision in hand, and so they’re the ones that matter. And then the field size at which its net present value is zero can be calculated with some simple trigonometry. In our example’s case, it’s 62 million barrels. So unless the well discovers at least 62 million barrels of oil, the discovery won’t be developed—which our geoscientists can tell us the likelihood of—and that represents our likelihood of a commercial, as opposed to only a geological, success.

Fig. 14.4—Using data points on a curve of field size vs. field values to approximate the probability of commercial success.

Back to our example. Now we can travel back in time again to the decision of whether or not to drill the exploration well knowing that we have a 75% chance that we’ll end up USD 18 million poorer because the exploration well doesn’t find oil, with the other 25% divided between: • A P90-sized discovery that would also leave us USD 18 million poorer because we wouldn’t develop it, • A P50-sized discovery that would have a value of the sum of the discounted net cash flows—discounted to when the exploration well is going to be drilled (because that’s the decision in hand now)—including the exploration well and, following on from it, development and production of the P50-sized discovery. Let’s say that’s USD 53 million. • A P10-sized discovery that would have a value calculated in the same way as the P50 but using the P10 outcome. Let’s say that’s USD 140 million. But how do we divide the remaining 25% chance between the possibilities of a P90-, P50-, and P10-sized discovery? The answer lies in why these three points are so convenient to use: because, with these three points, an off-the-shelf method exists to do so. That method is attributable to a gentleman, a geologist, called Roy Swanson, who demonstrated that a good approximation to an oil field’s mean average size can be derived by adding together 30% of the P90, 40% of the P50, and 30% of the P10. It’s called “Swanson’s rule.”

There are two caveats to using it to evaluate exploration opportunities where it’s also extended to similarly combine the values that are associated with the P90, P50 and P10 field sizes however. One is that the underlying distribution of possible field sizes must be no more than only moderately skewed (in other words, it shouldn’t have a long tail stretching out on only one side); and the second is that the shape of the distribution of a field’s value must be more or less similar to that relating to its size. The former may not be true if there exists a substantial, albeit improbable, upside to an exploration prospect, and the latter is unlikely to be true when fiscal terms behave disproportionately to field size; in other words they “tax” different field sizes differently (which is true of some production sharing contracts and even some concessions). If one or both of these situations apply to an exploration opportunity, then applying Swanson’s rule to their evaluation is flawed; and unless a lot more work is done to establish P90, P50, and P10 values of possible exploration outcomes, then it shouldn’t be used; and perhaps it’s best in that case to resort to a simpler analysis. In this chapter’s later examples and also in the chapter’s appendix (Appendix 14.1), I assume that both of the above assumptions hold true and that Swanson’s rule can therefore legitimately be used.

14.7 Decision Trees in One Minute Flat Exploration wells and similar situations where there are several possible outcomes from a decision, are usually shown in the form of decision trees. Decision trees map out all the possible consequences of a decision (see Fig. 14.5). In decision trees: • Squares represent decisions. So, on the left-hand side of the tree, we have a choice of whether to drill the exploration well. It’s up to us, we can either do it or not.

Fig. 14.5—Example of a decision tree.

• Circles represent uncertainty around a chance event. So, in the middle of the diagram, there’s uncertainty around the result of drilling the well; it might be a discovery or it might not be. And if it is a discovery, there’s also uncertainty around how big it will be. • Triangles represent possible outcomes. So, on the right-hand side of the diagram, there are, in our example, five possible outcomes. Stop the clock. Now back to our example again, whose decision tree looks like this (Fig. 14.6): I’ve added all the information we know to the decision tree, which is • The sum of all the discounted net cash flows that are associated with the whole journey from the very left to each possible outcome on the very right, discounted to when the exploration well is drilled, and • The likelihood of each result happening where there’s a chance event. All that remains to do now is condense all the possible outcomes of drilling the well (in other words, all the possible outcomes of the Yes branch of the tree) into its expected monetary value; which we do in exactly the same way as before, although the calculation has a few more parts now:

[75% × (−18)] + {25% × [(30% × 140) + (40% × 53) + (30% × (−18))]} As before, the calculation simply multiplies out the value of each possible outcome by the likelihood of it happening and then adds up all the individual results. And the answer is USD 1.0 million, which is encouragement to drill the well after all.

Fig. 14.6—Decision tree populated with our example’s information.

Tip: decision trees are constructed from left to right but are evaluated from right to left. The structure of the tree, which captures all the possible consequences of the decision, is mapped out forwards from the decision. The decision’s expected monetary value, however, is computed starting with the discounted cash flows associated with all its possible consequences and then working backward to calculate what, therefore, the decision should be.

14.8 Even More Realistically … Usually, much about an undiscovered oil field—especially its size—is extremely uncertain. Perhaps if it’s as small as we think it might be, then it would make us poorer to develop it, and so we wouldn’t. If it’s as big as we think it probably is, then we’d produce it through someone else’s production facility nearby, which would make us richer, but not much. And if it’s as big as we think it might be, then it would have its own dedicated production platform and make us richer beyond our wildest dreams.

But even after a discovery is made, much, if not all, of that uncertainty may remain intact. Often, therefore, we can’t possibly proceed to development and production having simply discovered some oil without first being much more certain about exactly what it is we’ve discovered. Without more certainty, we may choose to develop the discovery completely inappropriately by adopting a development approach that’s too conservative and doesn’t maximize its value or one that’s too speculative and might lose us money. So more often than not in the event of a discovery, further activity, including further drilling (of appraisal wells), is needed to provide more information about it and therefore give us more certainty. That comes at a cost, however, and so a separate decision has to be made about whether or not to incur it. The decision amounts to whether or not to buy more information to reduce our uncertainty. But we’ll only buy more information if it has some value; if it’ll pay for itself by enabling us to make more money in the long term (hence, the expression “value of information” that you might come across). This decision is informed in exactly the same way as the decision of whether to drill the exploration well. We use the appraisal well’s expected monetary value to establish whether its cost is worth incurring. Say we have an exploration target that looks like the one shown in Fig. 14.7. If the exploration well discovers oil, we (“we” being our geoscientists) believe the field will contain somewhere between 100 million barrels and a billion barrels of oil, and will most likely contain 350 million barrels. But we can reduce that uncertainty by drilling an appraisal well, in which case our decision tree looks like the one shown in Fig. 14.8.

Fig. 14.7—Example of exploration target.

Fig. 14.8—Decision tree including appraisal well information.

The decision relating to appraisal is coloured blue. If we drill an appraisal well and it finds oil as well, then it will prove the field can’t possibly be as small as 100 million barrels and is in fact at the higher end of the range we originally thought. If, however, we drill it and it doesn’t find oil, then it will prove the field can’t possibly be as big as 1 billion barrels and is in fact at the lower end of the range we originally thought. But in both cases, the uncertainty surrounding the size of the discovery reduces dramatically compared with not drilling the appraisal well. Here’s the appraisal part of the tree again, this time with all the information added to it that’s needed to calculate the appraisal well’s expected monetary value (Fig. 14.9): A few footnotes to the figure are: • This time, there are no scenarios in which we won’t develop and produce the discovery; we will, even in the low, P90, case. • The likelihood of the appraisal well finding oil is, once again, courtesy of our geoscientists. • The net present values are the sum of all the discounted net cash flows associated with the journey from the left to each possible outcome on the right,

discounted to when the appraisal well is drilled. • Although the high, P10, case is the same field size in two scenarios (one where the appraisal well is drilled and finds oil, and the other where it isn’t drilled) its value is higher in the scenario where the appraisal well is drilled because the discovery would be produced through a dedicated platform rather than, in the no-appraisal-well scenario, through someone else’s production facility nearby. In the latter case, without an appraisal well, it would be too risky to build a dedicated platform; we’d lose a great deal of money if the field did only contain 100 million barrels. Despite the higher cost, the former would produce the oil much faster, making it more valuable.

Fig. 14.9—Appraisal part of the decision tree populated with the information required to calculate the appraisal well’s expected monetary value.

The expected monetary value of the Yes (we drill the appraisal well) branch of the tree is then calculated as shown in Fig. 14.10: Which equals USD 2,130 million. And the expected monetary value of the No (we don’t drill it) branch is: (30% × 3,500) + (40% × 1,225) + (30% × 350), which equals USD 1,645 million So drilling the appraisal well increases the expected monetary value of the field’s

future and therefore we should do it.

Fig. 14.10—Expected monetary value of the Yes branch of the appraisal well.

Having made that decision, the No branch plays no further part in our evaluation. It becomes redundant, unless, that is, something happens—some new information materialises perhaps—to make us revisit it. Otherwise, we’ve made the decision to drill the appraisal well and the matter is closed. Now we can calculate the expected monetary value of the exploration well; and here’s the whole decision tree again (Fig. 14.11), this time populated with all the information we need to do so. Once again, a few footnotes: • Unless new information materialises that changes our minds, we will appraise a discovery, and so the No (we won’t) branch is redundant. • The likelihoods attached to chance events are sourced as before. • The net present values are the sum of all the discounted net cash flows associated with the journey from the very left to each possible outcome on the very right, discounted to when the exploration well is drilled (because that’s the decision in hand now). The values are therefore lower than we saw in the appraisal-only tree because (1) the exploration well occurs before the appraisal well, and so discounting has a greater impact on future cash flows; and (2) they include the cost of the exploration well. Then, the expected monetary value of the Yes (we drill the exploration well) branch of the tree is as follows, in Fig. 14.12.

Fig. 14.11—Decision tree populated with the information required to calculate the exploration well’s expected monetary value.

Fig. 14.12—Expected monetary value of drilling the exploration well.

The answer to which is USD 410 million. And, of course, not drilling the well—that is, doing nothing—has a value of zero. So drilling the exploration well has a higher expected monetary value than not drilling it and therefore we should drill it.

Of course, once the decision tree is established, some different calculations can be explored (forgive the pun) using it. For example, we can calculate what the minimum likelihood of a discovery is that results in a positive expected monetary value, which might be helpful if, for instance, our geologists are arguing over whether the likelihood of a discovery is 20% or 30% but calculations show that a 60% likelihood is needed to make the opportunity work—in which case the argument goes away. The above example is still relatively simple, however. Some especially large and complex fields might need three, four, or even more appraisal wells before a company is prepared to commit to development and production. And then decision trees, similarly, become large and complex (and specialist software is very helpful to solve them; several such products are available on the market). Here’s one last exploration target example (Fig. 14.13):

Fig. 14.13—A final example of an exploration target.

Our geoscientists have identified an exploration well location and, potentially, three appraisal well locations. So the opportunity’s decision tree looks like this (Fig. 14.14): It isn’t populated with any numbers and is the result of only the left-to-right pass that maps out the exploration well’s possible futures. For the last time, a few footnotes: • “Oil” and “No oil” are replaced with ticks and crosses, respectively, to make space. • Strictly speaking, the endpoints on the right-hand side (triangles) should be

circles—chance events—each with three possible outcomes: a P90 value, a P50, and a P10. But I’ve cheated to make the tree less busy. You can see straightaway that a lot of numbers are required to solve it, and those numbers require a lot of work to produce—and that’s without the very real possibility that plans might fundamentally change depending on drilling results (for example, different appraisal well locations might be chosen or the order the appraisal wells are drilled in might change, which would make for even more work). But set that work off against the tens of millions—and possibly hundreds of millions—of dollars at stake, and a lot of work is surely worth doing. Now, for fear of this chapter outstaying its welcome, it ends here. Later on, however, is an appendix that performs the right-to-left pass and calculates the expected monetary values of the appraisal wells and the exploration well, should you wish to see this last example out (see Appendix 14.1).

Fig. 14.14—Decision tree of the final example of an exploration target.

14.9 To Summarize Oil companies can convince themselves to spend what are sometimes vast sums of money drilling an exploration well without knowing whether they’ll find any oil or not by working out the prize they’ll win if they find some and the loss they’ll suffer if they don’t, and judging what they think the likelihoods are of those outcomes.

The potential prize and potential loss each multiplied by its likelihood of happening and then added together form the well’s expected monetary value. The expected monetary value is the average value of all the possible outcomes of a decision. So we should drill an exploration well if the expected monetary value of drilling it is greater than the value of not drilling it (which is zero, do nothing), and we shouldn’t drill it in the opposite circumstances. Usually, before an exploration well is drilled, oil companies are very uncertain about exactly how much oil a prospective field contains, and they still might not be that much wiser after a discovery is made. The more completely they can understand a potential field the better in order to make the most informed decision they can and maximize its value. Typically, therefore, oil companies determine three estimates of a prospective field’s size: its P90, its P50, and its P10—a low case, a mid-case, and a high case. These are convenient because in many situations they can be used in the proportions 30%, 40%, and 30%, respectively, according to Swanson’s rule, to approximate the field’s mean average size. Swanson’s rule can also be used to approximate the mean average field value using these low, mid- and high cases but the distributions of the field’s possible sizes and values must reasonably closely mirror each other. A helpful way of recognizing all the possible consequences of a particular decision, and also of hosting all the information required to calculate an expected monetary value, is the decision tree. Decision trees are constructed from left to right, the single trunk being the decision itself with a few, to several, to multiple branches dividing and separating from it that are all its possible consequences. Then expected monetary values are calculated from right to left, by: 1. Establishing whether potential outcomes would be developed and produced or not when the time comes and would make the company richer or poorer, 2. Establishing whether more information should be bought through appraising a discovery, and 3. Establishing whether the gamble of drilling the exploration well is worth taking. At each stage, only the relevant cash flows relating to, and forward of, the decision in hand count; and they’re discounted to when that decision is taken. Decision trees can be complicated, however, especially for large and/or complex fields where a discovery requires significant appraisal and other investigations to understand it fully; and in such cases, specialist software is available and helpful.

14.10 A Final Note This chapter has strayed into something that’s increasingly known as “decision analysis” in oil (and, indeed, other) companies. As the name suggests, decision analysis involves a more analytical approach to making decisions—or at least to informing them. Petroleum economics is at the heart of that, which is why I’ve

included at least the rudiments of it here. There are, however, much more sophisticated aspects of decision analysis that I haven’t mentioned for fear of straying too far from this book’s subject which, of course, is petroleum economics.

Chapter 15

Sensitivity Analysis Right from the start of this book I’ve diluted the belief a company can have that it will become “x” million dollars richer or poorer as a result of doing something new by prefixing it with the word “should.” And that’s because the only thing it can be certain of about a net present value, say, based on future cash flows is that it’s wrong. Given how much oil an oil field will produce is only a forecast, what the oil price will be when the oil is sold is only a prediction, and how much an oil field will cost to develop is only an estimate, it’s inconceivable that developing and producing a new discovery (for example) will make the company exactly as richer as the petroleum economist calculates. The upshot might even be that the company won’t be richer at all. So, often, the more interesting question is how wrong might that net present value be? Sensitivity analysis aims to answer that question by putting decision makers in the picture of how the company will fare if (or, rather, when) circumstances turn out differently. Ultimately, it aims to help decision makers make the right decisions. The information it provides can take a variety of forms, and not for the first time I’ve said this: individual companies have their own preferred forms, so be sure to understand your own company’s preferences. This chapter talks about the more common forms of sensitivity analysis: graphs, spider diagrams, and tornado charts. And it also introduces the ideas behind probabilistic petroleum economics.

15.1 Graphs The simplest form of sensitivity analysis is a graph—for example, of net present value (on the y-axis) and oil price (on the x-axis), or internal rate of return (on the y-axis) and the potential acquisition cost of an oil field the company is considering buying (on the x-axis). Graphs quite simply show how much net present value or internal rate of return (or the present value index) changes if something that determines it changes. They’re commonly used because of their straightforwardness to produce and easiness to understand, but they’re most appropriate to use when they explore a choice the company has rather than something over which it has little or no control. Take the two examples just mentioned, shown in Fig. 15.1.

Fig. 15.1—Examples of sensitivity analysis graphs.

The left-hand graph shows net present value vs. oil price relating to the development and production of a new discovery; and the right-hand graph shows internal rate of return vs. the potential acquisition cost of another company’s oil field. Both graphs provide information, but only one of them—the right-hand one—informs a choice that decision makers have. The left-hand graph provides information alright, but not information that presents decision makers with a choice (although gas price is occasionally a choice, of sorts, where sales are to state-owned organizations in certain countries and price is subject to negotiation), whereas the right-hand graph does—how much should we pay for the oil field? More to the point, however, if a separate graph is produced for every single forecast, prediction, and estimate that may turn out differently, there will be a plethora of graphs—and too many for decision makers to digest. So all such assumptions are usually better gathered together into a single graphical representation such as a spider diagram or tornado chart.

15.2 Spider Diagrams Spider diagrams are so-called because they look like spiders, sort of. Fig. 15.2 is one (though with 10 legs) relating to the development and production of a new discovery.

Fig. 15.2—Example of a spider diagram.

Spider diagrams explore the difference that percentage changes in assumptions make to, say, net present value (although it could equally be internal rate of return or present value index). What if the oil price is 10% higher than we’re assuming, what if production is 20% higher, what if it costs 5% less to develop the field, and so on. All the assumptions are conveniently gathered together, which not only makes for fewer graphs but instantly allows us to see what assumption the field’s net present value is most sensitive to percentage changes in (which is the one with the steepest gradient: oil price, in this case) and which assumption it’s the least sensitive to (which is the one with the gentlest gradient: operating costs in this case). So we can then spend more time thinking and talking about the more sensitive assumptions and less time on the less sensitive ones. And that’s the benefit of using spider diagrams. To their detriment, however, • Assumptions are explored one at a time, which may identify which has the most and which has the least impact (on net present value, say) given a certain change in percentage, but in reality they’re all likely to be different than we expect at the same time. • Decision makers have to form their own views regarding how wrong assumptions might be percentage-wise. Could our oil price prediction be wrong by 5%, or 10%, or 20%, or even more? Or our development cost estimates? And decision makers often aren’t experts in all these things to be able to do so. • Spider diagrams work only when assumptions are continuous rather than discrete—in other words, when they can adopt any value (within reason) rather than being restricted to one of only a few possibilities. All the assumptions in the above spider diagram are continuous because they can adopt any value within

the presented percentage range. But what about a situation in which an oil field may be subject to one of two branch remittance tax rates depending on the outcome of intergovernmental tax treaty negotiations? It isn’t easy to show something like that in a spider diagram.

15.3 Tornado Charts Tornado charts are so called because they look like tornados (sort of, because they’re usually wider at the top than at the bottom). Here’s one (Fig. 15.3) relating to the development and production of the same new discovery as before. Tornado charts are scenario based. Whereas spider diagrams explore the impacts of different percentage changes in the assumptions that determine, for example, net present value, tornado charts explore the impacts of an upside (being optimistic) and/or a downside (being pessimistic) for each assumption. The upsides and downsides that a tornado chart displays can be changes in percentages, like in a spider diagram, or, more helpfully, they can relate to particular scenarios. So tornado charts can present beliefs: the impact of what we believe the oil price might be, being optimistic and pessimistic, the impact of what we believe development costs might be, being optimistic and pessimistic, and so on; although being optimistic doesn’t mean every star aligns, and nor does pessimistic mean Armageddon. To be helpful to decision-makers, beliefs need to be realistically optimistic and realistically pessimistic.

Fig. 15.3—Example of a tornado chart.

In common with spider diagrams, all the assumptions in tornado charts are

conveniently gathered together, allowing us to instantly see which assumption affects, say, net present value the most if our optimism and pessimism are credible (which is the assumption with the widest bar) and which assumption affects it the least (which is the one with the narrowest bar). But they have the added advantage over spider diagrams that they can more easily deal with assumptions that are discrete in nature rather than continuous; for example, the middle bar shows what would happen if those intergovernmental tax treaty negotiations result in the lower of the two branch remittance tax rates. Like spider diagrams, however, assumptions are still explored one at a time, when in reality they’re all likely to be different than we expect at the same time. And whilst it’s possible to construct tornado charts that produce sensitivities in which more than one assumption at a time turns out better or worse than expected, they can only ever show a small sample of all the permutations that are possible. To reflect reality properly in that respect, we have to be more sophisticated.

15.4 Probabilistic Petroleum Economics Probabilistic petroleum economics aims to show every possible net present value, say, relating to whatever we’re evaluating and the relative probabilities of their occurrence given how certain—or uncertain—we are about the assumptions we’re making to calculate it (see Fig. 15.4).

Fig. 15.4—Probability density function of net present values.

And for decision-makers, wouldn’t that be just brilliant to see? Right in front of them is everything they want to know: the expected (mean) amount the company will

grow richer by; what happens if things go wrong, or if they go well; even the likelihood that the net present value will be less than, say, USD 100 million, although a different form of chart is needed for that, with the data plotted cumulatively (see Fig. 15.5).

Fig. 15.5—Cumulative frequency distribution of net present values.

But how are these curves generated? The answer, briefly, is as follows. 15.4.1 Generating Probability Curves. First, instead of using a particular number relating to something we can’t be certain about in our net present value (say) calculation, we accept the fact that we don’t know what the right number to use is and we use a range or a distribution of numbers instead. We admit to the calculation: “Look, I don’t know what’s going to happen and what the right number is, but I think it’ll be somewhere between here and here, though I’ve no idea where exactly.” Or, “Here are some possible values and the probabilities with which I think they’ll occur,” or “Here’s a P90, a P50, and a P10.” Then, secondly, we animate the calculation using, usually, Monte Carlo simulation. Monte Carlo simulation performs hundreds and even thousands of separate sensitivity analyses. It takes a random sample of every single forecast, prediction, and estimate that we’ve admitted we can’t be certain about and produces the resulting net present value (say). Or we can even tell the simulation not to take a random sample of a particular forecast, prediction, or estimate and instead correlate it with another one; that is, we can “link” it with something else if we think their behaviour is connected— for example, if the oil price goes up, drilling costs will too perhaps. And then the simulation does it again, and again, and again. It does it so many times that every possible combination of every single uncertainty is explored. All the resulting net present values are recorded and can then be compiled and

presented in whatever form is helpful—through graphs, charts, and/or statistics—to tell decision makers whatever they want to know about them. And that’s it. 15.4.2 More on the Two Steps. Needless to say, however, it isn’t as straightforward to carry out as these last few paragraphs might suggest. Think about those steps again. Step 1. On the first one: how do we capture our uncertainty relating to individual forecasts, predictions, and estimates? There are really only two ways: using relevant historical data or using the judgment of one or more experts, or perhaps a combination of both. If it’s something we’ve done plenty of times before or something there are published historical data on—economic data perhaps—then evidence exists of what might happen as opposed to our best guess of what we believe will happen, which we can use to represent our uncertainty. But that requires often time-consuming research and analysis to collect the data and establish whether it suggests any conclusions can be drawn regarding what, therefore, the range and shape of our uncertainty might be. Often, no such data exist, however, and so experts—if there are any, or otherwise whoever is most knowledgeable—have to be interrogated for their judgment of what might alternatively happen and what the likelihood of it is, which requires properly planned and time-consuming interviews. All this can take quite significant time to do and may still not represent our uncertainty accurately. Step 2. On the second step: any eight-year-old with a Meccano set will tell you that building a model that’s animated, that moves, is much harder than building one that doesn’t. Things happen that you don’t anticipate. Certain combinations of forecasts, predictions, and estimates might cause the model to break or errors to arise; and unless some sophisticated intelligence is built in—which in itself is dangerous because of the extra complexity it introduces—a model may still calculate a net present value using certain forecasts, predictions and estimates even if a particular combination of them isn’t possible. So these models have to be generally better engineered and quality assured.

15.5 On a Personal Note I’m an advocate of probabilistic petroleum economics and use it quite often. I like the honesty it offers me to not have to pretend I know something that I don’t, and so I can be up-front with the calculations and tell them only as much or little as I think I know. And then, in return, they tell me what all that means the answer might be and what faith I can have in any particular answer. If it’s done diligently, that is. But I’m also suspicious of it when I don’t know how the results have been generated, and decision makers should be too. It’s incredibly easy to generate a

distribution of possible net present values. You take each uncertainty, put an arbitrary range around its best forecast, prediction, or estimate—plus or minus twenty per cent will do—apply Monte Carlo simulation, and Bob’s your uncle: you have a distribution that can be made to look spectacularly professional and about which all sorts of highfalutin statistics can be quoted. But that’s worse than useless to decision makers because it may even mislead them into making the wrong decision. So decision makers, be very wary. Although I’m an advocate, however, I don’t use probabilistic petroleum economics all the time; I use it only when circumstances demand. Those circumstances are when a situation has more than only one or two material uncertainties. And that’s where, for example, tornado charts are helpful. They show how many material uncertainties there are by how many wide bars they have. If there are several of them, then that suggests to me I ought to use probabilistic petroleum economics. Otherwise, if circumstances are relatively certain—perhaps there are only a couple of material uncertainties if a field is fairly mature: exactly how much oil it will produce before it retires and what price the oil will sell for—then simpler ways may suffice to convey the spectrum of net present value possibilities. Perhaps a graph of net present value against oil price with different curves representing separate forecasts of how much oil remains would work perfectly satisfactorily. Where I do use probabilistic petroleum economics, I use the combination of Excel and Crystal Ball (I’ve already mentioned that I use Excel; and Crystal Ball provides both the ability to describe assumptions uncertainly and apply Monte Carlo simulation), although other products and combinations of them exist, and indeed many proprietary petroleum economics packages include a Monte Carlo simulation module. Lastly, I’ve included in the appendix to this chapter an indication of how I go about a probabilistic evaluation (Appendix 15.1). It’s a fact, though, that different evaluations throw up different uncertainties and ways of accommodating them, and that’s why I’ve referred to the appendix as an indication rather than an example.

15.6 To Summarize Sensitivity analysis addresses the question, “But what if everything doesn’t go according to plan?” The fact is everything never goes according to plan, and decision makers need to understand the consequences of that inevitability to help them make the right decisions. There are various ways of establishing and showing them the consequences: graphs, spider diagrams, tornado charts and using probabilistic petroleum economics, each of which takes more time and effort to produce than the last, but each of which provides more helpful information than the last, providing that —and this is crucial—they’re done diligently. If they aren’t, then sensitivity analysis is a waste of time and decision makers are in the dark about how their company’s investment in a project might fare. The degree of sophistication that petroleum economists should employ in

sensitivity analysis depends on circumstances. Where there are only one or two material uncertainties, then a simple graph may be perfectly adequate. If there are several, however, then to understand how much better or worse they may, all mixed up together, make a project’s net present value (say) calls for probabilistic petroleum economics to be used. Tornado charts are useful to establish just how many material uncertainties there are and therefore what degree of sophistication is appropriate to use.

Chapter 16

Acquisitions and Divestments So far when I’ve pointed to real situations in oil companies involving petroleum economics, I’ve used drilling an exploratory well or developing and producing a new discovery as examples. But another application of petroleum economics is when a company acquires (buys) another company’s oil field(s) or divests (sells) one or more of its own. Or even when it acquires or divests a whole company, which I’ll talk more specifically about later on. Whatever the application, however, petroleum economics is petroleum economics and the same laws apply. So everything discussed so far regarding relevant cash flows, net cash flows, and discounting applies in exactly the same way here. If we buy an oil field that’s already producing, that new production will provide new relevant cash flow into the company; or if it’s only a discovery, the cost of developing it will mean a new relevant cash flow out of the company. If we sell an oil field that’s already producing, the relevant cash flow that that production provides will go away, be lost; or if it’s only a discovery, the relevant cash flow relating to its development costs that would have gone out of the company now won’t. But there are up to two additional relevant cash flows that feature in acquisitions and divestments. One is the price the company pays to buy an oil field or the proceeds it receives if it sells one; and the other is—now brace yourself for some bad news—the difference that price or those proceeds make to how much tax the company pays. (If you want to take a moment to compose yourself after the upset of tax being mentioned again, please do.) The acquisition cost that a company incurs if it buys an oil field may be deductible for tax, and the proceeds it receives if it sells one may constitute taxable income. I say “may” because sometimes it is and it does, respectively, and sometimes it isn’t and it doesn’t; and now I’m going to repeat something I said many times in Chapter 4: go and see your by now good friends in the tax department if you’re involved in an acquisition or a divestment and ask them how the acquisition cost or disposal proceeds are treated for tax. I can, however, offer a few rules of thumb, which are these: • In a tax/royalty world, acquisition costs are never deductible for royalty or production taxes. They may, however, be deductible for corporate income tax (the profit-based tax that companies pay in the country in which the oil field resides). • Similarly in a tax/royalty world, divestment proceeds are never classified as

income for royalty or production taxes. They may, however, represent taxable income for corporate income tax. • In a production sharing or technical service contract world, neither acquisition costs nor divestment proceeds ever flow through the contract’s cost recovery and profit sharing mechanisms. They operate entirely outside the contract. • Nor, in a contractual world, are they relevant in the context of any regular taxes that are directly associated with the contract. Although acquisition costs and disposal proceeds may be outside the scope of the fiscal terms that petroleum economists ordinarily consider, there may nevertheless be other taxes that are specific to acquisition and divestment activity that do need to be considered. For example, selling an oil field may trigger a specific tax that applies to the amount received for it that isn’t related to any other taxes. Your tax department can tell you about these other taxes. It isn’t much to go on, I know, but it’s all the general guidance I can offer, I’m afraid.

16.1 The Acquisition and Divestment Process It’s worth saying a little bit about the process because not everyone has the opportunity to be involved in acquisitions and divestments. It works like this: 1. A data room is put together by the company selling the oil field, which, literally, it is a room full of data, containing every conceivable piece of information relating to the field being sold; every technical report, legal document, production sharing contract (if applicable), commercial agreement, and anything else that’s relevant (the room itself is often hidden away in a basement or other part of the building off the beaten track). The company is obliged to include all the information it’s aware of. If it doesn’t and some information comes to light later on that it knew about but hadn’t disclosed, there may be serious legal implications. 2. The selling company also, usually, puts together a flyer, which is a short description of the field meant to entice other companies to buy it (so it usually reads like an estate agent’s pitch for a house he’s selling—ideally situated, immaculately maintained, beautiful views, plenty of scope to extend, and so on). 3. Potential buyers then express an interest to participate in the process, although the selling company may choose only some of them to do so—those they believe are genuinely serious—if they need to control the number. The potential buyers then receive an information memorandum (IM for short; a brochure) containing more detailed information about the oil field for sale as a precursor to attending the data room. 4. Potential buyers are then usually allowed between 1 and 3 days in the data

room, depending on the nature of the field (and the generosity of the seller). 5. When potential buyers attend the data room, they’re usually first given a short management presentation to get them started, and then they’re given access to the information. Then they have to work like mad to understand the field as best and fast as they can. They may or may not be able to take copies of data and documents away with them (again, depending on the generosity of the seller, although also, sometimes, at the behest of government authorities who may not allow copies to be removed from the country). 6. After attending the data room, potential buyers usually then have between 1 and a few months to do all the work they have to do to come up with how much they’re prepared to offer, including getting their own management and board approvals to do so. During this time, the selling company remains available for questions and answers (Q&A) regarding anything that potential buyers still need clarifying. 7. The seller then receives all the offers, considers them, and notifies the winner that his offer is hereby accepted—and then a lot of champagne is drunk (although not necessarily by the people who did all the work). Increasingly, though, a lot of data rooms are held online, and physical data rooms, like the one I’ve just described, are replaced by an online site that shows potential buyers everything they need to see and allows them to examine the data in the comfort of their own offices. Or there might be a mixture of the two, where there’s a physical data room but certain information is available to see online. The petroleum economist’s job during this process, needless to say, is to derive the value—the net present value—of the oil field and whatever other measures are important to provide (internal rate of return in the buying company’s case, for example). The selling company’s petroleum economist derives what his company is prepared to sell it for, and the potential buying company’s what her company is prepared to pay for it. Both net present values (and any other measures) are anchored at a point in time called the “effective date,” which is the valuation date. It’s a particular day, so 1 January 2016, for example. In petroleum economics speak, it’s the discount date: the date at which the net present value (and other measures) is anchored (Chapter 8 talked about this). After a deal (a value, that is) has been agreed and all the champagne has been drunk, the petroleum economists’ jobs are done, although that isn’t the end of the process for the companies involved. Legal papers have to be finalised and other matters relating to the transfer of the field’s ownership from one company to another have to be executed, all of which takes time, often several months. So when the deal actually completes, there’s some balancing to be done regarding how much, say, income the selling company received after the effective date that really belonged to the buying company or how much money the selling company spent that really should have been spent by the buying company. All that is captured in a completion

statement, and the result of the balancing is either added to or deducted from the deal’s value as at the effective date; and that’s the sum of money that then actually changes hands. And then more champagne is drunk. Acquisitions and divestments are intensive for both the seller and the potential buyers, although especially for the latter (because the seller already knows and understands the field being sold). A great deal of work has to be done in a very short space of time. And given the short time scales, there’s often little time to accommodate formal reviews to check the work that’s been done. It’s often, therefore, a pressurised and stressful time, especially for petroleum economists, who are, of course, right at the heart of the valuation. I must say I’ve certainly had sleepless nights when I’m recommending that my company spends a billion dollars or more on buying an oil field. Am I sure I’ve captured everything? Am I sure I did that calculation correctly? Am I sure I took that into account properly? I certainly hope so. But acquisitions and divestments are usually exciting to be involved in too. Being part of a small team doing something that’s sometimes even company-changing is great to be a part of and is worth the sleepless nights.

16.2 Buying and Selling an Oil Field vs. Buying and Selling a Company A petroleum economist’s job is actually much the same no matter whether it’s an oil field that’s being bought or sold or a company. That’s because a company—an oil company, anyway—is largely worth what the sum of its oil fields are worth, which is the sum of their net present values. Certainly there are other assets and liabilities that are relevant as well, but usually they’re either straightforward to consider, such as debt (money owed to banks) and cash the company holds, or they’re modest in value terms, at least relative to what the company’s oil fields are worth, such as offices and employees (contrary to companies’ popular claim these days, employees aren’t in fact “our most valuable asset”). The one possible exception is if a company that another is considering acquiring owns taxable losses—that is, losses for tax purposes that it hasn’t been able to use and is carrying forward from year to year. If the acquiring company has positive taxable income itself and can get its hands on them, then those losses are worth money because they will reduce its tax bill. So, somewhat counterintuitively, a company may be prepared to pay a lot of money for another that hasn’t been very successful because—thanks to its failures—it has accumulated substantial taxable losses. So, by and large, calculate the net present values of a company’s oil fields and you’re most of the way to valuing the company. Calculating the value of another company is usually quite difficult in practice though because it’s generally the case that you don’t have much detailed information about their oil fields. When one

company is considering buying another one, the former doesn’t have automatic access to the latter’s data. That’s confidential. And there usually isn’t a data room; or if there is, its contents may be quite superficial because companies generally don’t want to be bought, and so they don’t have much incentive to cooperate. Often therefore, a company considering buying another one has to make do with whatever data it can get. It may already know the target company’s fields if it already owns an interest in one or more of them, but otherwise it has to glean information from wherever it can—from the other company if possible, from its annual reports, press reports, published technical reports, third parties (Wood Mackenzie, for example), rumours, gossip, anywhere; which is a dangerous game to play. A company buying another one usually doesn’t know exactly what it’s bought until it’s bought it, and it’s often the case that what it’s bought isn’t what it thought it was buying.

16.3 To Summarize The fundamental laws of petroleum economics are no different if a company is acquiring or divesting—buying or selling—an oil field than if it’s drilling an exploratory well or developing and producing a discovery. The task is still to identify all the associated, relevant cash flows and then derive and discount the net cash flow. Acquiring or divesting an oil field does give rise to up to two additional relevant cash flows, however. One is the acquisition cost (the cost of acquiring it) or the divestment proceeds (what the company receives for selling it) and the other relates to (feel free to insert an appropriate adjective of your own here) tax. Regarding the latter, an acquisition cost may be deductible against profit-based taxes, and divestment proceeds may constitute taxable income for the same taxes. Ask your tax department. Acquiring or divesting a company rather than an oil field still amounts to much the same exercise; the only difference is that there may be other assets besides the company’s oil fields that have to be factored into its value (e.g., debt, cash, offices, taxable losses). Ask your tax department about whether an acquisition cost or disposal proceeds relating to a company acquisition or disposal affects your company’s tax in any way. When an oil company sells an oil field, it provides potential buyers with all the information they need to value it—that is, to calculate its net present value and other associated measures—by placing the information in a physical and/or online data room. In the case of whole companies, however, potential buyers usually don’t have that luxury and instead have to beg, steal and borrow information (not literally; please don’t do anything illegal) from wherever they can.

Chapter 17

Lease vs. Purchase Decisions So far in this book’s discussions and examples of petroleum economics, the assumption has been that oil companies buy the production facilities—the platforms and so on—they need to develop and produce an oil field. There are occasions, however, when leasing them is a more attractive proposition.

17.1 What Is a Lease? A lease is a rental agreement that, on the face of it, is no different from the rental agreements you’re probably familiar with yourself. The word “rent” isn’t typically used in the oil industry, however; “lease” is. For all intents and purposes, though, they mean the same thing. One party to the agreement builds or buys something and is its owner but allows another party to use it in return for periodic payments. The two words—“rent” and “lease”—do imply a slightly different arrangement when they’re applied in certain situations, to property, for example, where a rental is a short-term, relatively flexible contract between owner and user whereas a lease is a longer-term, relatively less flexible one. And I do use the word “rent” in this chapter, but only in the general sense of referring to the periodic payments that are made to an asset’s owner. In a lease, the party that builds or buys the asset and is its owner is called the lessor and the party who uses it and makes periodic payments to the lessor is called the lessee. So, for example, in my dreams, Aston Martin is the lessor of a brand-new DBS (a white one), and I’m the lessee.

17.2 Why Lease? Companies (and people) lease something for a variety of reasons. Perhaps: • They neither have nor are able to raise the funds to buy it. • It’s simply more convenient, whatever the numbers say; and the shorter the duration an asset is needed for, the more convenient it is. You or I wouldn’t go to the trouble of buying a house, registering it, fitting it out, insuring it, and so on, and then reselling it if we only wanted it for a week’s holiday; we’d rent one instead. Similarly, oil companies wouldn’t go to the trouble of buying, or even more trouble building, say, a floating production, storage, and offloading vessel (FPSO) if they only needed one for 6 months; they’d lease one instead. • Cancellation options in the lease may be valuable to an oil company if it’s possible that an oil field may be shorter-lived than expected. If the company

didn’t need whatever-it-is for as long as they thought they would, the lessor would take it back and re-lease it to someone else, whereas if they’d bought it, the company would be left with equipment that, perhaps, it has no further use for and can’t sell, or would lose money on its resale. • It’s cheaper for another party to pay for it to be built or bought (and maybe even maintained and operated as well) and then to lease it from them rather than pay for it ourselves in the first place. The first three points don’t really warrant further discussion, and so this chapter concentrates on the last one. But how can it be cheaper for someone else to build or buy something and then we lease it from them rather than we pay for it ourselves in the first place? Well, if the subject of the lease is, say, a car, then a specialist car-leasing company may be able to buy it more cheaply than you or I can because the company buys a lot of cars and so can negotiate substantial discounts with manufacturers. Even after the leasing company’s own costs and profit are included in the rent, it may still work out cheaper for us to lease it from them. But that situation doesn’t typically apply in the oil industry. The sort of equipment oil companies use isn’t off the shelf or a production line; it’s typically designed and built specially for a particular oil field. So, chances are, the same price will apply to any buyer. It’s still possible for someone else—a lessor—to buy that equipment more cheaply than an oil company can, however; but how is more subtle: • The lessor may have access to cheaper finance than the oil company; in other words, the lessor can borrow money at a lower interest rate than the oil company can and is therefore able to build or buy the equipment more cheaply. • The lessor may be subject to a higher tax rate than the oil company, and so the tax depreciation it can claim on the capital expenditure is more valuable. • The lessor may be in a position to claim the aforementioned tax deductions immediately, whereas the oil company may not be (if it isn’t paying tax because it has taxable losses that it’s carrying forward from year to year). My apologies for “tax” making an almost certainly unwelcome appearance again, but it couldn’t be helped, I’m afraid.

17.3 More on What Is a Lease In the oil industry a lease has up to two elements: a capital element and an operating element. The capital element relates to building or buying the asset, and the operating element relates to its operation and maintenance. To put it another way, the capital element is the alternative to the oil company building or buying the asset itself, and the operating element is the alternative to it operating and maintaining the asset itself. A lease may contain both or just the capital element (if an oil company uses

another company to operate and maintain, say, an FPSO it owns, that wouldn’t be called a lease). Lease payments, be they capital or operating in nature, are entered as operating costs in net cash flow computations, and both tax/royalty and contractual fiscal terms recognise them as such. But beware; they have a dark secret that means they can’t be directly compared with the alternative capital expenditure, and I’ll come back to that later on.

17.4 Evaluating a Lease vs. a Purchase Whether an oil company buys, say, an FPSO or leases it, it ends up with the same FPSO. Buying an asset or leasing it has nothing to do with the asset itself; it’s about how it’s paid for. Here’s a conversation I recall. The scene is my office; the telephone rings. I exchange some small-talk (and some gossip) with the person I know on the other end of the line, and then the conversation—about a discovery and potential field development in the US Gulf of Mexico—goes like this: Other person:

There are three ways we can develop this field. Either we tie it back to a third-party [production] facility, or we develop it stand alone (independently), or we use a leased [production] facility.

Me:

No, there aren’t. There are two ways you can develop it: you can tie it back to a thirdparty facility, or you can develop it standalone.

Other person: Me:

Why can’t I lease? Maybe you can, but leasing isn’t a wholly different way of developing the field. It’s just a different way of paying for how you intend to develop it anyway.

In a situation like this, there are potentially two decisions to make: • An investment decision relating to whether or not we want to invest in the opportunity at all, and • A financing decision relating to how we’re going to pay for the production facilities and equipment that are needed if we do want to. I say “potentially” because not all assets can, in practice, be leased and therefore there is no financing decision to make if they can’t be. An asset can usually be leased providing a market exists for its reuse when a lessee has no further use for it. So a mobile asset that can easily be picked up and put somewhere else, such as an FPSO, can usually be leased, whereas a fixed platform or a subsea pipeline usually can’t be. Where both decisions—investment and financing—do apply, they are separate and sequential.

17.4.1 To Invest or Not to Invest? In the earlier conversation’s example, the investment decision concerns whether we want to develop the field at all; is there a way of developing it that represents a sufficiently attractive investment? And if there’s more than one way, which is the most attractive? To make this decision, we apply petroleum economics in exactly the way we’ve talked about it leading up to this chapter. If an option is to use an FPSO, then we assume we build or buy it and the cash flows associated with it are the capital expenditure to do so. Let’s say that we do indeed have an attractive investment on our hands, and its most attractive option is to use an FPSO. The investment decision is therefore made and that box is ticked. 17.4.2 To Lease or to Buy? Now we have a financing decision we may wish to make. But why have I said we only “may wish” to make? Do you recall a little earlier (in 17.2) when I listed some possible ways a lessor might be able to build or buy the equipment that an oil company needs more cheaply than the oil company can? Briefly, the lessor might be able to borrow money more cheaply than the oil company, or they might have a higher tax rate, or they might be able to make use of tax depreciation faster. But what if none of these situations apply? What if an oil company has a rockbottom borrowing rate, there’s no tax rate difference, and it pays tax and therefore can make use of the tax depreciation itself immediately? And what if as well, to address the other reasons why a lease may appeal to an oil company, it’s perfectly able to raise the funds if it wanted to, it needs the asset for a long time, and it’s confident that the oil field will produce for at least ten years (so cancellation options would have little or no value)? Well, then, it’s difficult to see how leasing will benefit the oil company. And the fact is those conditions are often the case, which is why leasing isn’t especially common and why oil companies often don’t even explore it. But let’s say here that, for whatever reason, an oil company thinks it may be beneficial to lease, say, an FPSO. How does it go about choosing whether to lease it or to buy it? What Not to Do. First, let me say what mustn’t be done, because it’s a mistake that’s commonly made, and it can be a very expensive one. What mustn’t be done is to populate two different cash flow models, one with all the data associated with the lease and the other with all the data associated with the purchase (e.g., production profiles, product prices, capital and operating costs), compute their net cash flows, and see which one, once discounted at the company’s hurdle rate, has the greater net present value. That mustn’t be done because the lease costs in the lease option and the capital expenditure in the purchase option aren’t equivalent. Implicit in the lease costs is the lessor’s cost of borrowing the money it needs to

build the FPSO, which is passed on to the lessee to pay. But the alternative capital expenditure includes no such oil company’s cost to borrow the money with which to build it itself. And that’s the lease costs’ dark secret that I alluded to earlier. (I’ve seen this mistake made several times, and certainly in one case it resulted in a company taking on a lease with a staggering, credit-card-like equivalent cost of borrowing, which cost that company very dearly indeed.) What to Do. So this anomaly has to be remedied before the two alternatives can be properly compared. Either the lessor’s borrowing costs have to be removed from the lease, or the oil company’s borrowing costs have to be added to the purchase; the latter being preferable because we don’t know what the lessor’s borrowing costs are. In big picture terms then, what this means is the oil company borrows something in both options. In the lease, it borrows the asset from the lessor, pays the lessor rent until it’s finished with the asset, and then gives the lessor its asset back. In the purchase, it borrows money from the bank; builds or buys the asset with it, pays the bank interest until it’s finished with the asset, then sells it and gives the bank its money back. Either way, the oil company borrows something, pays money from time to time to whoever it’s borrowed from, and then gives back again what it borrowed. The question, then, is who is it cheaper to borrow from, the lessor or the bank? To answer that, we have to calculate the total costs of obtaining the asset from both of them.

17.5 Financing Costs Financing costs are the total cost of obtaining an asset. • For a lease, they include not only the lease costs but also their impact on any other cash flows—for example, cost recovery in a production sharing contract and the benefits that come about through their deduction for tax. • For a purchase, they include the capital expenditure (to buy the asset), the company’s borrowing costs, and, the same as for a lease, their impact on any other cash flows. And they also include any other cash flows that would be different if the asset is purchased rather than leased, so maintenance and operating costs if they’re included on the lease side. Financing costs include every cash flow whereby the lease and purchase options differ, and every cash flow is excluded that would be exactly the same whether the asset is leased or purchased. Let’s go straight into a simple example, which is whether to lease or purchase an FPSO—and in this example, only the capital elements of leasing and purchasing are compared to keep matters simple. Numbers are all USD million, except Row 14. Year Year Year Year Year Year Year Year Year Year Year Year Year

1

2

1 Lease an FPSO 2 Tax savings 3 Lease cost after tax, undiscounted 4 Lease cost after tax, discounted

(482)

3

4

5

6

7

8

9

10

11

12

(82)

(82)

(82)

(82)

(82)

(82)

(82)

(82)

(82)

(82)

25

25

25

25

25

25

25

25

25

25

(57)

(57)

(57)

(57)

(57)

(57)

(57)

(57)

(57)

(57)

(54)

(53)

(51)

(50)

(49)

(47)

(46)

(45)

(44)

(42)

5 Borrow money to build FPSO

750

6 Pay for FPSO

(750)

7 Tax savings from FPSO

45

45

45

45

45

8 Interest on loan

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

9 Tax savings from interest

9

9

9

9

9

9

9

9

9

9

9

9

13

10 Repay loan

(750)

11 Sell FPSO

150

12 Purchase cost after tax, undiscounted

24

24

24

24

24

(21)

(21)

(21)

(21)

(21)

(21)

(21) (600)

13 Purchase cost (435) after tax, discounted

24

23

23

22

21

(18)

(18)

(17)

(17)

(16)

(16)

(15) (431)

14 Discount factors

1.000 1.028 1.057 1.086 1.117 1.148 1.180 1.213 1.247 1.282 1.318 1.355 1.393

Before we draw any conclusions, we’ll walk down the numbered rows in the table; the lease option first: Line 1: the annual cost of leasing the FPSO, assuming USD 225,000 per day and 365 days per year. The numbers are in brackets, meaning they’re cash flows out of the company; they’re costs, in other words. Line 2: tax savings from spending the lease costs, assuming a 30% tax rate. Lease costs are expensed (deducted in full, immediately) for tax, and so we save 30% × USD 82 million in tax, which is USD 25 million. If that shortcut calculation doesn’t resonate with you, think about it the long way. Let’s say without the lease costs we’d have USD 100 million of taxable income from our other oil fields. So, given a 30% tax

rate, we’d pay USD 30 million of tax. Now, with the lease costs, we can deduct USD 82 million from the USD 100 million, leaving USD 18 million of taxable income, on which we’d pay USD 5 million of tax (30% of USD 18 million). So we’ve saved the difference between USD 30 million of tax and USD 5 million, which is USD 25 million. Whichever method of calculation you use, the USD 25 million is, effectively, cash flow into the company because it’s money (tax) saved, and so it’s positive. Line 3: the annual lease costs after tax (–82 + 25). Line 4: the annual lease costs after tax, discounted, which I’ll come back to when we get to Line 14. And now the purchase option: Line 5: the bank lends us USD 750 million, which is cash flow into the company and therefore positive. Line 6: the money from the loan is instantly turned around again and used to build an FPSO, which is cash flow out of the company and therefore negative. Line 7: tax savings from building the FPSO, assuming the capital expenditure is depreciated on a straight line over 5 years (so in Year 1, one-fifth of USD 750 million, multiplied by 30%—the tax rate—equals USD 45 million, which is, effectively, cash flow into the company because it’s tax money saved, and so it’s positive). Line 8: interest on the loan, assuming a 4% interest rate (and is cash flow out of the company). Line 9: tax savings as a result of paying interest, assuming that the interest is tax deductible (30% of USD 30 million equals USD 9 million of cash flow into the company). Line 10: repayment of the loan, which is cash flow out and therefore negative. Line 11: the FPSO is sold, which is cash flow in and therefore positive. Line 12: the annual, total costs, after tax, of buying the vessel, which is the sum of Lines 5 to 11. Line 13: the annual, total costs, after tax, of buying the vessel, discounted (which is discussed below). Now let’s stop and draw breath for a moment. What we’ve achieved in Lines 3 and 12 is to isolate the cash flows that are purely and simply associated with leasing and buying the FPSO. Line 3 represents what we’d actually pay, all things considered, to

borrow it from the lessor; and Line 12 represents what we’d actually pay, all things considered, to borrow money from the bank, build our own FPSO, and then sell it again when we’ve finished with it. What we need to do now is establish which cash flow represents the best deal; which we can do by discounting the cash flows, although in a slightly different way from the discounting we’ve done so far. Given that we have two costs of borrowing (the asset from the lessor and money from the bank), the discount rate we use relates to exactly that—our cost of borrowing, the borrowing-from-the-bank part of our weighted average cost of capital (WACC). So, returning to the table, Line 14 contains discount factors that use the same 4% interest rate from Line 8, expressed after tax, because that’s our true cost of borrowing once the interest we pay is deducted for tax (again, assuming it’s deductible for tax, of course). Year 1’s discount factor is simply 1 (assuming we apply beginning-year discounting). Then, subsequent years are the preceding years’ factors multiplied by 1 + [4% × (1 − 30%)], where 4% is the interest rate which (1 − 30%) converts to after tax if the tax rate is 30%. Finally, having discounted the two cash flows, we can conclude that we should purchase the vessel because to do so costs less (it has a lower net present value, USD 435 million as opposed to USD 482 million to lease it). That was a simple example, however, in more ways than one. In real life, the question of whether to lease or purchase can be rather more complicated, but I’m going to banish that discussion to an appendix (Appendix 17.1).

17.6 After the Financing Decision Is Made Once we’ve established whether to lease or purchase is the right thing to do, that’s that question answered. After that, the cash flows in the project’s net cash flow computations assume that outcome. So if the answer is to lease, then the computations from now on assume all the lease-related cash flows, and if the answer is to purchase, then they assume all the purchase-related ones from now on (although excluding the cash flows relating to financing because that responsibility is handed back to the hurdle discount rate).

17.7 Where to Go for Help I spent a lot of time in Chapter 4 encouraging you to go and talk to your tax department whenever you’re in any doubt about anything tax related (and even, frankly, when you’re not in any doubt). Regarding lease or purchase decisions, the equivalent department is your treasury team. A company’s treasury is the custodian of its corporate finances. Therefore, any borrowing commitment the company accepts should be blessed by the experts that work in it, and they should be included in any lease or purchase evaluation. In my experience, treasury people are just as nice as tax people.

17.8 A Closing Remark Like many other topics covered in this book, different companies go about evaluating whether to lease or purchase an asset slightly differently, notwithstanding that leasing is relatively uncommon. I’ve chosen to discuss the topic as I have, however, because I believe it provides a logical foundation to understanding the principles that are involved; and so I hope it helps you to more easily understand your own company’s approach if it’s slightly different.

17.9 To Summarize A lease is a rental agreement whereby one party builds or buys something—an asset —and is the owner of it but allows another party to use it in return for periodic payments. There are sensible reasons to lease an asset and reasons that are nonsense: • The overarching sensible reason is that it makes financial and business sense to do so. • The reasons that aren’t sensible are that leasing preserves capital or that it provides a means of circumventing capital budgets or hiding liabilities (or at least making them less obvious) in company reports. Of the reasons that aren’t sensible, leasing doesn’t preserve capital at all because a company can equally borrow the money to purchase an asset as borrow the same asset directly from a lessor. And regarding using leasing to circumvent capital budgets or hide liabilities, these are dangerous practices that, if allowed to drive decisions, can leave companies significantly worse off financially. Any asset can usually be leased providing a market exists for its reuse when a lessee has no further use for it. So, for example, an FPSO can usually be leased but a fixed platform or a subsea pipeline usually can’t be. Before even considering whether to lease or purchase an asset, a company must decide whether an opportunity deserves investment in the first place. Only once that hurdle is cleared should a company’s mind turn to the financing decision of whether to borrow an asset from a lessor or borrow money from a bank to build or buy it itself. Financing decisions are made by establishing the total equivalent costs of leasing an asset and, alternatively, purchasing it, with the latter including the oil company’s costs associated with borrowing the money to do so (because the lease includes the lessor’s equivalent costs), and then discounting both costs at the oil company’s aftertax cost of borrowing. The option with the lower discounted cost is the winner.

Chapter 18

Presenting Your Work So you’ve done some quite magnificent petroleum economics work: you’ve established relevant cash flows, you’ve modeled a particularly complex production sharing contract, you’ve computed net present values and other results, you’ve explored whether your company should lease assets or buy them, and you’ve built decision trees. Now you have to explain your work to other people, tell them your conclusions, and propose your recommendations. This chapter isn’t concerned with the art of standing up in front of (sometimes important and slightly scary) people and speaking. That’s a part of presenting your work of course, but it’s beyond the scope of this book’s purpose. All I will say about that is the more often you do it—and seek constructive feedback afterwards—the better and more comfortable it feels; whereas do it only once or twice a year, and the worse and more uncomfortable it feels. Instead, this chapter is concerned with the material you have to produce in report form to be read or in PowerPoint to be presented, because if there’s one weakness that many technical professionals (and I include petroleum economists in this category) can be accused of, it’s not doing themselves and their work justice when it comes to the material they report or present. Because of that material’s quality relative to the work that’s been done, I’ve witnessed audiences perceive that very good work is poor and, even worse, very poor work is outstanding. My message certainly isn’t “Don’t worry about the work, worry about the presentation;” it’s that its presentation matters just as much as the work itself does. So don’t just produce the first material that comes to mind. Give ample thought and time to designing and creating it, and certainly don’t leave it to the last minute and throw it together. Here are a few guidelines I try and abide by.

18.1 Know Your Audience Who’s the report or presentation material for? Is it your peers or your boss or your senior managers? That’s important because one size doesn’t fit all here. All these people have different agendas, ranging from the former’s to understand whether your work is sound, to the latter’s disregard of that (if your peers and boss have endorsed it anyway) and focus on what decision they should therefore make. It’s true. Explaining the detail of the calculations you’ve built and the numbers you’ve generated to an audience of senior managers will probably see you fired, not promoted; and presenting far-reaching recommendations to your peers probably won’t earn you an invitation to go to the pub with them after work.

So my advice is to detach yourself from what you’ve done, clear your head, ask yourself earnestly what your audience wants and needs to know about your work, and tell them as relevantly as you can.

18.2 Pictures Speak Louder Than Words Petroleum economics is a quantitative subject, and therefore there’s inevitably a need for tables of numbers, charts, and graphs in reports and presentations. In many cases, companies have standard forms of these that must be adopted, at least for formal submissions to senior management and the board of directors. In other cases, however, petroleum economists have some freedom, which I encourage you to embrace. Be creative. Think about how information can most helpfully be presented both to support your point and to help your audience see it. The right chart or graph can deliver a killer blow, and it’s certainly worth looking for. To counterbalance what I’ve just said, however, don’t make charts and graphs more complicated than you have to. Don’t get carried away; a particular chart may be ingenious, but if it’s unfathomable, then it’s useless. Be creative, yes, but also keep it simple. And therein lies the challenge, which is by no means easy to accomplish.

18.3 Read it, Reread it, and Then Read it Again You probably remember your English teacher (or your own language’s teacher) at school (like I do) telling you that writing is important, and thinking (like I did) that it doesn’t matter because “I want to do a job that involves numbers, not words.” Well, it turns out that he or she was right. Writing is important, very important. When someone reads something you’ve written, you want him or her to concentrate on and understand what you’re saying, not to have to rebuild sentences to make them make sense first, or be distracted by grammar and spelling mistakes, or be left behind by language they’ve never heard (I could probably go on). You simply can’t be an excellent petroleum economist if you can’t write such that your reader receives your message loud and clear.

18.4 To Summarize Don’t be fooled into thinking that because this chapter is quite short it can’t be very important. Wrong. The importance of communicating your thoughts, ideas, results, conclusions, and recommendations can’t be understated. It’s crucial that you carry out the task thoughtfully and carefully because the material you produce that other people will see can make or break a piece of work—your work. Whether it’s a report or a slide-show, think about the audience it’s intended for and address only what the people in it want and need to know. Think hard about what

charts and graphs support your message the clearest. And write properly good (!)

Chapter 19

Going About Your Business Petroleum economics is a unique discipline. It puts you at the very heart of appraising the (sometimes very significant) opportunities that your company is considering investing (sometimes extraordinarily large amounts of money) in. That in turn puts you right in the middle of the many other disciplines—petroleum engineering, cost engineering, commercial, and so on—that supply the raw material for that appraisal: information. Without information, petroleum economists are helpless. So, clearly, they must be adept at collecting it. And that requires possession of a very different set of skills than those discussed so far in this book. Not skills relating to the science of petroleum economics, but those relating to the art of dealing with people. And that art is the subject of this chapter. Let me say straightaway, however, that I’m no psychologist, and so I’m not going to talk about the theory of this. I couldn’t even if I wanted to. Rather, I’m going to talk about my own observations and conclusions from practicing the art of dealing with people over the years and watching others practice it, both the good and the bad.

19.1 The Challenge The principal challenge of dealing with people is that it involves dealing with people, and it can be challenging for all sorts of reasons: • Seniority. You may have to deal with people who are much more senior than you, which can be daunting. • Age and experience. You may have to deal with people who are older and (they, and indeed you may think) wiser than you (and sometimes they are, but not always). • Subject matter. As a petroleum economist you need to collect information from many other disciplines, some of which you may understand and some you may not at all. • Culture. Approaching people with different cultural and perhaps other backgrounds to yours may feel uncomfortable. Last, and perhaps most important, is personality. You may have to deal with people you don’t much care for and/or who, despite your best efforts to achieve the contrary, don’t care for you much either. But deal with them you must. And sometimes not only deal with them but negotiate with them and persuade them as well. The challenge isn’t to be underestimated.

19.2 The Prize The prize of overcoming the challenge is a valuable one, however, and is, potentially, even a priceless one when “deal with (people)” takes the form of “talk to (people).” The difference between emailing someone to ask for information and properly engaging in a conversation with them regarding your request can be enormous. The former will result in an answer; the latter, however, will result in a discussion, and discussions are much more helpful than answers. They educate, they clarify (what you’re asking for and why), and they share responsibility for the work in hand, all of which reduces the risk of error. And not only that, they build relationships as well, which is a definite bonus.

19.3 The Sixty-Four-Thousand-Dollar Question The sixty-four-thousand-dollar question, of course, is, so how do you deal with people? How do you go about engaging them so that they provide the information you need, when you need it? Truthfully, I don’t know the answer, but here are my own dos and don’ts that may not amount to the whole sixty-four-thousand-dollars’ worth, but they do represent a substantial deposit towards it. Do: • Go and see people if you can (if they’re in the same building, for example); failing that, telephone them rather than email them and have that discussion rather than request information remotely, clinically, and facelessly. • Be brave if the thought of approaching someone in person or telephoning them is daunting. It’s never as bad as you think it’s going to be—or perhaps I should say rarely as bad! • Be clear about what information you want, but be open to discussing it. • Be clear, as well, about when you need the information (your deadline) and what format you’d prefer it to be provided in. • If you don’t understand something, ask. People are usually very glad to explain things. • Treat the information you’re given confidentially (unless there’s good reason not to). If people feel the information they’ve provided is misused or miscommunicated, they may not provide you with any more. • Always but always be respectful and courteous, and be grateful for people’s help. Don’t: • Do the opposite of any of the above dos! • Make unreasonable demands of people. Urgent requests are sometimes

unavoidable, but otherwise make your request in good time. • Name-drop; tell them that Dr. Frankenstein (who they will know is the top boss in Paris) has asked you to do this work—the inference being that if they don’t cooperate, it’s her they’ll have to answer to. Name-dropping doesn’t endear you to people and I don’t recommend it. • Go above people’s heads if they aren’t immediately helpful. There’s usually a good reason why they aren’t (they’re too busy with other things); so discuss your request with them and if it can’t be satisfied, then agree to escalate it to your respective managers for them to find a solution. • Be secretive about your results unless there’s a particular need to be. People are usually interested in the role their information plays in what you’re doing; so show them, and then they’ll feel they’re getting something in return. I’ve often thought it surprising that this aspect of a petroleum economist’s work isn’t generally recognised as much as, say, technical and presentational skills are. The skill of dealing with people is just as important. Good petroleum economists have it, and a common reason why some petroleum economists aren’t as good is because they don’t.

19.4 To Summarize You can’t be a petroleum economist—at least, a very good one—working in isolation, sitting behind your desk. It isn’t that kind of job. It’s the kind of job that involves—no, requires—contact with other people; the people who provide all the information petroleum economists need to produce their net present values and other results. Although that contact is potentially challenging in a variety of ways, it’s vital that petroleum economists rise to the challenge because the information they seek to gain from it is their lifeblood; they simply can’t function without it. How petroleum economists go about collecting the information they need to do their jobs determines the amount and quality of what they collect, however, and the more engagement they have with the information’s suppliers, the better.

Chapter 20

Some Advice for Non-professional Petroleum Economists Many, if not most, if not the vast majority of petroleum economist practitioners come from other disciplines and backgrounds. Few are petroleum economists by original training. Therefore, as non-professional petroleum economists, they face an immediate challenge, aside from being or becoming proficient at petroleum economics. That challenge is the same one that managers in companies and even prime ministers and presidents face, which is that they have to surrender their particular cares and interests of the past and now also care about much else besides. Managers have to care about what happens in parts of the company they may never have worked in, and prime ministers and presidents have to care about what happens all over their country and not just in one part or aspect of it anymore. In the same way, nonprofessional petroleum economists have to care about other disciplines now in addition to their own original one. I mention this because my experience is that petroleum economists who come from other disciplines don’t always do that. Petroleum engineers doubling as petroleum economists tend to give much consideration to oil and gas production profiles but less to commercial issues; and accountants doubling as petroleum economists tend to give much consideration to financial matters but less to technical ones; and so on. And that won’t do. A net cash flow computation is only as good as its weakest component. The production profile that’s used may well be inch-perfect, but that doesn’t count for much if the gas price has been over-simplistically calculated or the company’s tax position hasn’t been properly researched. That isn’t to say, however, that petroleum economists should be experts at everything; naturally, they can’t be. They can’t speak every language that’s spoken in an oil company—petroleum engineering, finance, tax, and so on—all fluently. But they do need to be able to speak them all up to a point, at least to the level a tourist needs to survive abroad and be able to say the equivalent of “Please can I have a table by the window?” in the language of petroleum engineering and “Is there a pub near here?” in geology. And that requires effort and determination. Some introductory courses in the various disciplines may help, as may books. Often, though, the most useful education is to go and talk to people. So if you’re a petroleum engineer, go and talk to your commercial colleagues and find out about the ins and outs of their world; or if you’re part of the commercial team, go and talk to

your technical colleagues. It’s the best training there is, guaranteed.

20.1 To Summarize Many practitioners of petroleum economics aren’t petroleum economists by original training. A challenge such non-professional petroleum economists face, therefore, is to resist the pull of their own discipline and branch out and engage with other disciplines as well. Petroleum economics necessarily involves different disciplines, and the absence of that broad engagement in evaluations risks its failure.

Chapter 21

A Final Word If someone had told me many years ago that I’ll spend the greater part of my career as a petroleum economist, I’d have given them a very blank look because, back then, I had no idea what a petroleum economist was. But if, instead, they’d said I’ll spend the greater part of my career wrestling with some real and significant intellectual and analytical challenges in helping to inform sometimes multibillion-dollar decisions working alongside just about every professional discipline under the sun in an industry that makes the world go round, my eyes would have lit up. That’s the world petroleum economists live in. The role of a petroleum economist is to establish whether the decision before an oil company to do something, such as drill an exploratory well, should make it richer or poorer, and this book has been about the principles and processes petroleum economists use to do that. Six out of the first seven chapters were about cash flow because understanding what cash flows will be created or affected by doing that something is at the heart of the task. So petroleum economists must establish all the resulting “relevant” cash flows—by estimation or calculation—relating to the something (development of a discovery, for example) and then compute its “net” cash flow. Building up a picture, a qualitative model, of the project showing what and where the field is, what it produces, where it sells what it produces and how those products get to where they’re sold can help establish those relevant cash flows. But be careful because some cash flows are more transparent than others. Perhaps there are commercial complexities involving, say, a carry (where, for example, one partner company in an oil field agrees to fund another’s development costs and then the latter repays the former from its share of production once the field is producing, with interest). But also take comfort that someone (a cost engineer perhaps) or something (a commercial agreement perhaps) can explain each and every relevant cash flow. It’s just a question of finding that person or thing. Then we talked about “discounting”, which serves to recognise that an oil company’s projects also have to contribute to the non-project-specific costs that the company as a whole has to bear—its costs of capital, corporate costs, and unsuccessful exploration—by stripping them out of projects’ net cash flows. What’s then left behind is a project’s net present value, which is the actual amount the project, or decision to pursue it, should make the company richer or poorer by. As well as net present value, we also talked about some other ways of measuring whether projects are good, bad, or indifferent in the context of being an attractive investment, which were “internal rate of return” and “present value index.” These

three measures all tell the story differently, and their stories should all be listened to. The next few chapters discussed how petroleum economics deals with matters such as inflation—which introduced the ideas of “real” and “nominal” terms—and also what happens when, as is inevitable in the global oil industry, cash flows come and go in all sorts of different currencies. And to a great extent, that was the end of talking about the science of petroleum economics, and from there we moved on to its practice, both in a technical sense—its application to exploration, acquisitions and divestments, leasing vs. purchasing assets, and the need, sometimes, to take the difference between the before and after in order to get to the right answer, which is “incremental economics”—and in a softer sense: the behaviours that help petroleum economists get the job done and what to think about when it comes to presenting their work and results. And in the middle of all that, we acknowledged the sad but inevitable truth that the answer petroleum economists come up with is bound to be wrong, but, on the bright side, how they can use sensitivity analysis to provide a helpful sense to decision makers of how wrong the answer might be. These principles and processes are universal; they can be applied anywhere a company asks the question, “Should doing this make us richer or poorer?” Finally, let’s not forget we also learned that thinking about tax makes you afraid of the dark and that geologists like drinking beer, which may or may not be helpful to know. I leave that to your judgment. In all seriousness, though, I hope you’ve enjoyed reading this book, because I very much wanted it to be as much of a pleasure to read as it’s possible for a book like this to be. If I’d thought there wasn’t going to be any pleasure in reading it, I’d rather not have written it. For my part, however, it remains the book that I wish I could have read twenty years ago.

Appendices Appendix 4.1.1—Concession fiscal terms: only one oil field inside the ring fence and therefore no consolidation. Year Year Year Year Year Year Year Year Year Year 1 2 3 4 5 6 7 8 9 10

Year 11

Production, millions of barrels of oil Oil production from the oil field.

3

8

10

10

8

6

4

2

1

Oil price, USD per barrel Price the oil is sold for.

90

90

90

90

90

90

90

90

90

270

720 900 900 720 540 360

180

90

A Revenue generated, USD million

Year 12

Production multiplied by oil price; so USD 270 million in year 3 is 3 million barrels of oil × USD 90 per barrel. B Exploration and appraisal costs, USD million

30

Exploration and appraisal costs associated with the oil field. C Development capital costs, USD million

250

300

150

50

50

The costs incurred to develop it. D Production operating costs, USD million

50

50

50

50

50

50

50

All the costs associated with producing and transporting the oil to where it’s sold. E Decommissioning costs, USD million

100

The cost of decommissioning the oil field’s wells and production facilities F Royalty, USD million

27

72

90

90

72

Let’s say there’s a royalty which is 10%, so 10% of revenue (Row A) Corporate Income Tax calculation: This is a profit-based tax levied on the company that owns the oil field. Development capital depreciation for tax: Development capital costs incurred in year 2

50

50

50

50

50

Development capital costs incurred in year 3

60

60

60

60

60

54

36

18

9

Development capital costs incurred in year 4

30

G Total development capital depreciation for tax

110

30

30

30

30

140 140 140 140

30

Let’s say that development capital costs are depreciated on a straight line basis over 5 years (so in equal amounts over five years) starting only when the associated assets are placed into service; and let’s say that the assets associated with the capital spent in Years 2 and 3 become operational when production begins and that the assets associated with the capital spent in Year 4 become operational immediately. Then the USD 250 million spent in Year 2 is depreciated as USD 50 million per year for 5 years starting in Year 3, the USD 300 million spent in Year 3 is depreciated as USD 60 million per year for 5 years also starting in Year 3, and the USD 150 million spent in Year 4 is depreciated as USD 30 million per year for 5 years starting in Year 4. H Revenue

270

720 900 900 720 540 360

180

90

The oil has been sold on the open market and therefore the value of what we’ve received for tax purposes is the same as Row A. Deduct exploration & appraisal costs (from Row B)

30

Deduct development capital depreciation (from Row G)

110

140 140 140 140

30

Deduct production operating costs (from Row D)

50

50

50

50

50

50

50

50

50

Deduct royalty (from Row F)

27

72

90

90

72

54

36

18

9 100

Deduct decommissioning costs (from Row E) I Total deductions

30

187

262 280 280 262 134

86

68

59

100

Exploration and appraisal costs are expensed (deducted in full and immediately), as are production operating costs, royalty and decommissioning costs; and development capital costs are depreciated according to the section ending with Row G. J Taxable losses carried forward

(30) (30)

K New additions to taxable income (30)

83

458 620 620 458 406 274

112

31 (100)

L Taxable income in the year

53

458 620 620 458 406 274

112

31

M Taxable losses for the year

(30) (30)

(100)

Row J contains the losses, if any, carried forward from the previous year. Being the first year, there’s no entry in Year 1. Row K calculates the taxable income that has been newly generated in the current year, which is revenue (Row H) minus total deductions (Row I). So in Year 1 there’s no revenue; however there are total deductions of USD 30 million (arising from exploration and appraisal costs), meaning there are new additions to taxable income for the year of negative USD 30 million. In Year 3, there’s revenue of USD 270 million and deductions of USD 187 million (made up from USD 110 million of development capital depreciation, expensed production operating costs of USD 50 million and royalty of USD 27 million) meaning there are new additions to taxable income for the year of 270 less 187 equals USD 83 million. Row L calculates taxable income in the year and is the sum of Row J (losses, if any, carried forward from the previous year) and Row K (new additions to taxable income in the year). but only when the result is greater than zero, otherwise it’s zero. Row M calculates whether there are any losses to carry forward to the following year by adding together Rows J and K and subtracting Row L. Note that we have USD 100 million of taxable losses left at the end

of Year 12. Note: numbers in brackets are negative and those without brackets are positive. N Tax to pay ignoring (for a moment) taxable losses from decommissioning

27

229

310

310 229 203 137

56

16

Let’s say the tax rate is 50%, so, in each year, the tax to pay is 50% of that year’s taxable income (Row L). O Tax effects of carrying back losses from decommissioning

(50)

Let’s assume that the country we’re operating in allows taxable losses relating to decommissioning costs to be carried back indefinitely. Then, providing there has been sufficient taxable income in prior years (in Row L) to cover the losses we want to carry back, we can deduct those losses against that prior years’ income. So of the USD 100 million of decommissioning costs, USD 31 million can be deducted against prior Year 11, and the remainder (USD 69 million) can be deducted against prior Year 10; in which case we should have paid no tax in Year 11 and only 50% (the tax rate) of 112 (Year 10’s taxable income) minus 69 (the rest of the decommissioning costs not deducted in Year 11), which is USD 22 million. So we should have paid USD 16 million less tax in Year 11 and USD 34 million less in Year 10 (56 we did pay minus 22 we should have paid), which is USD 50 million in total, and which is the same as 50% (the tax rate) multiplied by USD 100 million of decommissioning costs. However, we can’t of course turn back the clock and actually pay less tax in earlier years, and so the USD 50 million of tax that we overpaid is reimbursed in Year 12 - although in practice it may come later after all the administration is complete. Net cash flow Revenue (from Row A)

270

720 900 900 720 540 360

180

90

Exploration & appraisal costs (from (30) Row B) Development capital costs (from Row C)

(250) (300) (150)

Production operating costs (from Row D)

(50) (50) (50) (50) (50) (50) (50)

(50)

(50)

Royalty (from Row F)

(27) (72) (90) (90) (72) (54) (36)

(18)

(9)

Decommissioning costs (from Row E) Tax (from Rows N and O) Net cash flow

(100) (27) (229) (310) (310) (229) (203) (137) (56) (30) (250) (134) 219 450 450 369 233 137

56

(16)

50

16

(50)

Note in this table that amounts are designated positive or negative (denoted by brackets) depending on whether they represent positive cash flow (into the company) or negative cash flow (out of the company). So in Years 1. 2. 3 and 12 there’s net negative cash flow (out of the company) and in all the other years there’s net positive cash flow (into it). Appendix 4.1.2 – Concession fiscal terms: more than one oil field inside the ring fence and therefore consolidation. Year Year Year Year Year 5 Year Year Year Year Year 1 2 3 4 6 7 8 9 10 Production, millions of barrels of oil Oil production from the oil field.

3

8

10

10

8

6

4

2

Year 11 1

Year 12

Oil price, USD per barrel Price the oil is sold for.

90

90

90

90

90

90

90

90

90

A Revenue generated, USD million

270

720

900

900 720 540 360

180

90

Production multiplied by oil price; so USD 270 million in Year 3 is 3 million barrels of oil × USD 90 per barrel. B Exploration and appraisal costs, USD million

30

Exploration and appraisal costs associated with the oil field. C Development capital costs, USD million

250

300

150

50

50

The costs incurred to develop it. D Production operating costs, USD million

50

50

50

50

50

50

50

All the costs associated with producing and transporting the oil to where it’s sold. E Decommissioning costs, USD million

100

The cost of decommissioning the oil field’s wells and production facilities F Royalty, USD million

27

72

90

90

72

54

36

18

9

Let’s say there’s a royalty which is 10%, so 10% of revenue (Row A) Corporate Income Tax calculation: This is a profit-based tax levied on the company that owns the oil field. Development capital depreciation for tax: Development capital costs incurred in Year 2

50

50

50

50

50

Development capital costs incurred in Year 3

60

60

60

60

60

30

30

30

30

30

140

140

140 140

30

Development capital costs incurred in Year 4 G Total development capital depreciation for tax

110

Let’s say that development capital costs are depreciated on a straight line basis over 5 Years (so in equal amounts over 5 years) starting only when the associated assets are placed into service; and let’s say that the assets associated with the capital spent in Years 2 and 3 become operational when production begins and that the assets associated with the capital spent in Year 4 become operational immediately. Then the USD 250 million spent in Year 2 is depreciated as USD 50 million per year for 5 Years starting in Year 3, the USD 300 million spent in Year 3 is depreciated as USD 60 million per year for 5 Years also starting in Year 3, and the USD 150 million spent in Year 4 is depreciated as USD 30 million per year for 5 Years starting in Year 4. H Revenue

270

720

900

900 720 540 360

180

90

The oil has been sold on the open market and therefore the value of what we’ve received for tax purposes is the same as Row A. Deduct exploration and appraisal costs (from Row B)

30

Deduct development capital depreciation (from Row G)

110

140

140

140 140

30

Deduct production operating costs (from Row D)

50

50

50

50

50

50

50

50

50

Deduct royalty (from Row F)

27

72

90

90

72

54

36

18

9 100

Deduct decommissioning costs (from Row E) I Total deductions

30

187

262

280

280 262 134

86

68

59

100

Exploration and appraisal costs are expensed (deducted in full and immediately), as are production operating costs, royalty and decommissioning costs; and development capital costs are depreciated according to the section ending with Row G. J Taxable income in the year

(30)

83

458

620

620 458 406 274

112

31

(100)

Taxable income in the year is revenue (Row H) minus total deductions (Row I). That leads, in Years 1 and 12, to negative taxable income. However, we assume in this example that the company generates taxable income from its other oil fields inside the ring fence that is more positive than the taxable income for this oil field is negative, such that once they are consolidated the company’s overall taxable income in the year is still positive. So the negative taxable income in Years 1 and 12 isn’t negative in an absolute sense for the company, it just means that its taxable income in those years is lower than it would have been without this oil field (and higher in the other years). Note: numbers in brackets are negative and those without brackets are positive. K Tax to pay

(15)

42

229

310

310 229 203 137

56

16

(50)

Let’s say the tax rate is 50%. So, in each year, the tax to pay is 50% of that year’s taxable income (Row J), and because there’s negative taxable income in some years of Row J, there is also negative tax to pay in those years. But in the same way that taxable income isn’t negative for the company as a whole (because we assume it has positive taxable income generated by its other oil fields inside the ring fence that more than makes up for this field’s negative taxable income), the tax it has to pay isn’t negative for the company as a whole either. The negative tax represents how much less tax it would pay in that year with this oil field than without it. Imagine if the company’s consolidated taxable income without this oil field was USD 150 million in Year 1. In that case, it would pay USD 75 million in tax (50% × USD 150 million). Now with this oil field, its taxable income in Year 1 is USD 120 million (150 minus 30); and the tax it would pay is USD 60 million (50% × USD 120 million). So it would pay USD 15 million less tax, which is the same number as in Year 1 of Row K. Net cash flow Revenue (from Row A) Exploration and appraisal costs (from Row B)

270

720

900

900 720 540 360

180

90

(50) (50) (50) (50) (50)

(50)

(50)

(30)

Development capital costs (from Row C)

(250) (300) (150)

Production operating costs (from Row D)

(50) (50)

Royalty (from Row F)

(27) (72)

(90) (90) (72) (54) (36)

(18)

(9)

Decommissioning costs (from Row E) L Tax (from Row K) Net cash flow

(100) 15

(42) (229) (1310) (310) (229) (203) (137) (56)

(15) (250) (149) 219

450

450 369 233 137 56

(16)

50

16

(50)

Note in this table that amounts are designated positive or negative (denoted by brackets) depending on whether they represent positive cash flow (into the company) or negative cash flow (out of the company). So in Years 1, 2, 3 and 12 there’s net negative cash flow (out of the company) and in all the other years there’s net positive cash flow (into it). Also note that the tax saving we talked about for Year 1 (in the second paragraph under Row K) has come through as positive USD 15 million (in Row L), so cash flow into the company; which happens because this project enables the company to spend USD 15 million less (in tax), which is the same as saying that it’s provided USD 15 million of positive cash flow in that year. Appendix 4.2.1 – Production sharing contract: development capital is drip-fed into the cost pool, contractor is liable for decommissioning costs, and contractor pays tax. Year Year Year Year Year Year Year Year Year Year 1 2 3 4 5 6 7 8 9 10 Production, millions of barrels of oil Oil production from the oil field.

Year 11

3

8

10

10

8

6

4

2

1

A Oil price, USD per barrel Price the oil is sold for.

90

90

90

90

90

90

90

90

90

B Revenue generated, USD million

270

720 900 900 720 540 360

180

90

Year 12

Production multiplied by oil price; so USD 270 million in Year 3 is 3 million barrels of oil × USD 90 per barrel. C Exploration and appraisal costs, USD million

30

Exploration and appraisal costs associated with the oil field. D Development capital costs, USD million

250

300

150

50

50

The costs incurred to develop it. E Production operating costs, USD million

50

50

50

50

50

50

50

All the costs associated with producing and transporting the oil to where it’s sold. F Decommissioning costs, USD million

100

The cost of decommissioning the oil field’s wells and production facilities G Royalty, USD million

27

72

90

90

72

54

36

18

Let’s say there’s a royalty, which is 10%, so 10% of revenue (Row B) Production Sharina Contract calculations, all USD million unless otherwise stated:

9

H Production available for Contractor’s cost recovery and profit

243

648 810 810 648 486 324

162

81

Royalty is subtracted from the oil field’s revenue before the Production Sharing Contract’s cost oil and profit oil calculations; so revenue generated (Row B) minus, royalty (Row G). Although this is ‘production’ available for Contractor’s cost recovery and profit, it’s measured here in USD million. In order to do these calculations, either production has to be converted into money (as is done here) or else money has to be converted into production (the Contractor spends money, not barrels of oil) by dividing it by the oil price (so, for example, USD 90 million equates to 1 million barrels of oil if the oil price is USD 90/bbl). The calculations can be done in either ‘currency’. Cost oil: Development capital additions to the cost pool from: Development capital costs incurred in Year 2

50

50

50

50

50

Development capital costs incurred in Year 3

60

60

60

60

60

30

30

30

30

30

140 140 140 140

30

Development capital costs incurred in Year 4 I Total development capital additions to the cost pool:

110

Let’s say development capital costs are added to the cost pool in equal amounts over 5 years starting only when the associated assets are placed into service; and let’s say the assets associated with the capital spent in Years 2 and 3 become operational when production begins and the assets associated with the capital spent in Year 4 become operational immediately. Then the USD 250 million spent in Year 2 is added as USD 50 million per year for 5 years starting in Year 3, the USD 300 million spent in Year 3 is added as USD 60 million per year for 5 years also starting in Year 3, and the USD 150 million spent in Year 4 is added as USD 30 million per year for 5 years starting in Year 4. PSCs tend to be silent regarding what happens if the contract terminates before certain capital can be added to the cost pool in full - for example if capital is spent in this project when the contract has only three more years to run. Unless the contract says otherwise, one should assume that a proportion of that capital doesn’t reach the cost pool and therefore isn’t recovered, which of course has implications for Contractors’ appetite to spend capital when the contract is nearing the end of its life. J Decommissioning cost additions to the cost pool

20

20

20

20

20

Decommissioning costs aren’t always explicitly mentioned in Production Sharing Contracts and, if they aren’t, decommissioning is assumed to be the responsibility of governments because, usually, assets (e.g., wells, production facilities, pipelines) become their property either immediately or once the Contractor has recovered its costs of them. Here, let’s say the Contract stipulates that decommissioning costs are the Contractor’s responsibility and they’re added to the cost pool in 5 equal amounts over the last 5 years of production from the oil field (if the Contract does specify that decommissioning is the Contractor’s responsibility, it will explain how they’re recovered). Opening balance of cost pool (see note 1 below)

30

Additions to the cost pool from: Exploration and appraisal costs

30

30

(from Row C) Development capital costs (from Row I)

110

140 140 140 140

30

Production operating costs (from Row E)

50

50

50

50

50

50

50

20

20

20

20

20

160

190 190 190 210 100

70

70

70

190

190 190 190 210 100

70

70

70

Decommissioning costs (from Row J) Total additions to the cost pool

30

K Contractor’s cost recovery (see note 2 below) Closing balance of cost pool (see note 3 below)

30

50

50

30

Note 1: The opening balance of the cost pool in any year is its closing balance in the previous year. It’s automatically zero in the contract’s first year. Note 2: The Contractor’s cost recovery is the lesser of a) the production available for Contractor’s cost recovery and profit (Row H) and b) the cost pool’s balance, which is the sum of its opening balance plus the total additions for the year (in other words all the costs the Contractor has spent but hasn’t recovered yet). If the production available for Contractor’s cost recovery and profit is less than what’s in the cost pool, then cost recovery is limited to that amount. If, however, the available production is more than what’s in the cost pool, then all of it can be recovered. A useful check to confirm the integrity of these calculations is to compare the total costs incurred and total cost recovery, which ought to be the same (here, they’re both USD 1,280 million). If they aren’t the same (and the calculations are correct) then perhaps some costs aren’t being fully recovered. Note 3: The closing balance of the cost pool in any year is the opening balance for that year, plus the total additions for the year, less the amount that has been recovered by the Contractor. Profit oil: Total Profit Oil

53

458 620 620 438 386 254

92

11

The production available for Contractor’s cost recovery and profit that’s left over after the Contractor’s cost recovery, so Row H less Row K. L Contractor’s share of total Profit Oil

13

115

155 155 110

97

64

23

3

Let’s assume the Contractor receives 25% of the total Profit Oil (although, as said in Chapter 4, the mechanism can be much more complicated than that). M Contractor’s revenue entitlement

203

305 345 345 320 197 134

93

73

The sum of Contractor’s cost recovery (Row K) and Contractor’s share of total Profit Oil (Row L). Contractor’s production entitlement in millions of barrels

2.26 3.38 3.83 3.83 3.55 2.18 1.48 1.03

0.81

Contractor’s revenue entitlement (Row M) divided by oil price (Row A) equals the volume of oil that the Contractor is entitled to take and sell. Tax calculations: Revenue (from Row M)

203

305 345 345 320 197 134

93

73

The oil has been sold on the open market and therefore this is the value of what the Contractor has received for tax purposes. Deduct exploration and appraisal costs (from Row C)

30

Deduct development capital depreciation (from Row I)

110

140 140 140 140

30

Deduct production operating costs (from Row E)

50

50

50

50

50

50

50

20

20

20

20

20

190 190 190 210 100

70

70

70

50

50

Deduct decommissioning costs (from Row J) N Total deductions

30

160

Exploration and appraisal costs are expensed (deducted in full and immediately), as are production operating costs, and, in this example, development capital costs and decommissioning costs are depreciated in accordance with their treatment for cost recovery. Royalty isn’t a deduction for tax because it isn’t paid for out of the Contractor’s entitlement revenue; it’s taken by the government ‘off the top’, meaning before the PSC’s remuneration mechanisms are applied. O Losses carried forward

(30) (30)

P New additions to taxable income (30)

43

115

155 155 110

97

64

23

3

Q Taxable income in the year

13

115

155 155 110

97

64

23

3

R Losses to be carried forward

(30) (30)

Row O contains the losses, if any, carried forward from the previous year. Being the first year, there’s no entry in Year 1. Row P calculates the taxable income that has been newly generated in the current year, which is revenue (Row M) minus total deductions (Row N). So in Year 1 there’s no revenue; however there are total deductions of USD 30 million (arising from exploration and appraisal costs), meaning there are new additions to taxable income for the year of negative USD 30 million. In Year 3, there’s revenue of USD 203 million and deductions of USD 160 million (made up from USD 110 million of development capital depreciation and expensed production operating costs of USD 50 million) meaning there are new additions to taxable income for the year of 203 less 160 equals USD 43 million. Row Q is the sum of Row O (losses, if any, carried forward from the previous year) and Row P (new additions to taxable income in the year), but only when the result is greater than zero. Row R calculates whether there are any losses remaining to carry forward to the following year by adding together Rows O and P and subtracting Row Q. Note: numbers in brackets are negative and those without brackets are positive. S Tax to pay

5

46

62

62

44

39

25

9

1

Let’s say the tax rate is 40%, so, in each year, the tax to pay is 40% of that year’s taxable income (Row Q). Net cash flow Revenue (from Row M) Exploration and appraisal costs (from Row C) Development capital costs

203

305 345 345 320 197 134

(30) (250) (300) (150)

93

73

(from Row D) Production operating costs (from Row E)

(50) (50) (50) (50) (50) (50) (50)

(50)

(50)

Decommissioning costs (from Row F)

(100) (5)

Tax (from Row S) Net cash flow

(30) (250) (152)

(46) (62) (62) (44) (39) (25) 59

233 233 226 108

58

(9)

(1)

34

22

(100)

Note in this table that amounts are designated positive or negative (denoted by brackets) depending on whether they represent positive cash flow (into the company) or negative cash flow (out of the company). So in Years 1, 2, 3 and 12 there’s net negative cash flow (out of the company) and in all the other years there’s net positive cash flow (into it). Appendix 4.2.2 – Production sharing contract: a cost recovery ceiling exists, contractor is liable for decommissioning costs, and tax is paid on behalf of the contractor by the government. Year Year Year Year Year Year Year Year Year Year 1 2 3 4 5 6 7 8 9 10 Production, millions of barrels of oil Oil production from the oil field

Year 11

3

8

10

10

8

6

4

2

1

A Oil price, USD per barrel Price the oil is sold for.

90

90

90

90

90

90

90

90

90

B Revenue generated, USD million

270

720

900 900 720 540 360

180

90

Year 12

Production multiplied by oil price; so USD 270 million in Year 3 is 3 million barrels of oil × USD 90 per barrel. C Exploration and appraisal costs, USD million

30

Exploration and appraisal costs associated with the oil field. D Development capital costs, USD million

250

300

150

50

50

The costs incurred to develop it. E Production operating costs, USD million

50

50

50

50

50

50

50

All the costs associated with producing and transporting the oil to where it’s sold. F Decommissioning costs, USD million

100

The cost of decommissioning the oil field’s wells and production facilities G Royalty, USD million

27

72

90

90

72

54

36

18

Let’s say there’s a royalty which is 10%, so 10% of revenue (Row B) Production Sharina Contract calculations, all USD million unless otherwise stated:

9

H Production available for Contractor’s cost recovery and profit

243

648

810 810 648 486 324

162

81

Royalty is subtracted from the oil field’s revenue before the production sharing contract’s cost oil and profit oil calculations; so revenue generated (Row B) minus royalty (Row G). Although this is production’ available for contractor’s cost recovery and profit, it’s measured here in USD million. In order to do these calculations, either production has to be converted into money (as is done here) or else money has to be converted into production (the contractor spends money, not barrels of oil) by dividing it by the oil price (so, for example, USD 90 million equates to 1 million barrels of oil if the oil price is USD 90/bbl). The calculations can be done in either ‘currency’. I Maximum production available for Contractor’s cost recovery

97

259

324 324 259 194 130

65

32

Let’s say that only 40% of production is made available for the purpose of Contractor’s cost recovery, in other words there’s a cost recovery ceiling of 40%, so 40% of Row H. Cost oil: Development capital additions to the cost pool from: Development capital costs incurred in Year 2

250

Development capital costs incurred in Year 3

300 150

Development capital costs incurred in Year 4 J Total development capital additions to the cost pool:

550

150

Let’s say development capital costs are added to the cost pool only when the associated assets are placed into service; and let’s say the assets associated with the capital spent in Years 2 and 3 become operational when production begins and the assets associated with the capital spent in Year 4 become operational immediately. Note that, in this example, these costs join the cost pool as a single amount and aren’t drip-fed into it. K Decommissioning cost additions to the cost pool

20

20

20

20

20

Decommissioning costs aren’t always explicitly mentioned in production sharing contracts and, if they aren’t, decommissioning is assumed to be the responsibility of governments because, usually, assets (e.g., wells, production facilities, pipelines, etc) become their property either immediately or once the contractor has recovered its costs of them. Here, let’s say the contract stipulates that decommissioning costs are the Contractor’s responsibility and they’re added to the cost pool in five equal amounts over the last 5 years of production from the oil field (if the Contract does specify that decommissioning is the contractor’s responsibility, it will explain how they’re recovered). Opening balance of cost pool (see Note 1 below)

30

533

Development capital costs (from Row J)

550

150

Production operating costs

50

50

Additions to the cost pool from: Exploration and appraisal costs (from Row C)

30

474 200

5

30

50

50

50

50

50

50

50

43

(from Row E) Decommissioning costs (from Row K) Total additions to the cost pool

30

L Contractor’s cost recovery (see note 2 below) Closing balance of cost pool (see note 3 below)

30

30

20

20

20

20

20

600

200

50

50

70

70

70

70

70

97

259

324 250

70

70

70

65

32

533

474

200

5

43

43

Note 1: The opening balance of the cost pool in any year is its closing balance in the previous year. It’s automatically zero in the contract’s first year. Note 2: The contractor’s cost recovery is the lesser of a) the maximum production available for contractor’s cost recovery (Row I) and b) the cost pool’s balance, which is the sum of its opening balance plus the total additions for the year (in other words all the costs the contractor has spent but hasn’t recovered yet). If the maximum production available for contractor’s cost recovery is less than what’s in the cost pool, then cost recovery is limited to that amount. If, however, the maximum available production is more than what’s in the cost pool, then all of it can be recovered. A useful check to confirm the integrity of these calculations is to compare the total costs incurred and total cost recovery, which ought to be the same. Here, however, the Contractor has spent USD 1,280 million but only recovers USD 1,237 million because some costs (late in the oil field’s life) aren’t being fully recovered. Note 3: The closing balance of the cost pool in any year is the opening balance for that year, plus the total additions for the year, less the amount that has been recovered by the contractor. Excess Cost Oil

74

189 124

60

Because, here, a certain proportion of the production available for contractor’s cost recovery and profit is being allocated to contractor’s cost recovery, there will be periods when that’s too much, and then the question is what does the Contract do with the surplus? Usually what it does is treat it as if it was profit oil - divided between the Contractor and the government - sometimes in exactly the same way profit oil is divided, although sometimes differently M Contractor’s share of Excess Cost Oil

19

47

31

15

Let’s assume that Excess Cost Oil is treated in exactly the same way as Profit Oil (and the Contractor receives 25% of it). Profit oil: Total Profit Oil

146

389

486 486 389 292 194

97

49

If 40% of the production available for contractor’s cost recovery and profit has been allocated to contractor’s cost recovery, then 60% is available for Profit Oil, so 60% of Row H. N Contractor’s share of total profit oil

36

97

122 122

97

73

49

24

12

Let’s assume the contractor receives 25% of the total Profit Oil (although, as said in Chapter 4, the mechanism can be much more complicated than that). Tax calculations: O Revenue

134

356

446 390 215 174 134

89

45

The sum of the contractor’s cost recovery (Row L), contractor’s share of Excess Cost Oil (Row M) and the contractor’s share of total Profit Oil (Row N).

The oil has been sold on the open market and therefore this is the value of what the Contractor has received for tax purposes. Development capital depreciation for tax: Development capital costs incurred in Year 2

50

50

50

50

50

Development capital costs incurred in Year 3

60

60

60

60

60

30

30

30

30

30

140

140 140 140

30

Development capital costs incurred in Year 4 P Total development capital depreciation for tax

110

Let’s say that development capital costs are depreciated for tax on a straight line basis over 5 Years (so in equal amounts over 5 Years) starting only when the associated assets are placed into service. Then the USD 250 million spent in Year 2 is depreciated as USD 50 million per year for 5 Years starting in Year 3, the USD 300 million spent in Year 3 is depreciated as USD 60 million per year for 5 Years also starting in Year 3, and the USD 150 million spent in Year 4 is depreciated as USD 30 million per year for 5 years starting in Year 4. Deduct exploration and appraisal costs (Row C)

30

Deduct development capital depreciation (Row P)

110

140

140 140 140

30

Deduct production operating costs (Row E)

50

50

50

50

50

50

50

50

20

20

20

20

20

190 190 210 100

70

70

70

50

Deduct decommissioning costs (Row K) Q Total deductions

30

160

190

Exploration and appraisal costs are expensed (deducted in full and immediately), as are production operating costs; development capital costs are depreciated as described immediately above, and decommissioning costs are depreciated in accordance with their treatment for cost recovery. Royalty isn’t a deduction for tax because it isn’t paid for out of the contractor’s entitlement revenue; it’s taken by the government ‘off the top’, meaning before the PSC’s remuneration mechanisms are applied. R Losses carried forward

(30) (30) (56)

S New additions to taxable income (30)

(26)

T Taxable income in the year U Losses to be carried forward

(30) (30) (56)

(25)

166

256 200

5

74

64

19

110

256 200

5

74

64

19

(25)

(25)

(25)

Row R contains the losses, if any, carried forward from the previous year. Being the first year, there is no entry in Year 1. Row S calculates the taxable income that has been newly generated in the current year, which is revenue (Row O) minus total deductions (Row Q). So in Year 1 there’s no revenue; however there are total deductions of USD 30 million (arising from exploration and appraisal costs), meaning there are new additions to taxable income for the year of negative USD 30 million. In Year 4, there’s revenue of USD 356 million and deductions of USD 190 million (made up from USD 140 million of development capital depreciation and expensed production operating costs of USD 50 million) meaning there are new additions to taxable income for the year of 356 less 190 equals USD 166 million.

Row T is the sum of Row R (losses, if any, carried forward from the previous year) and Row S (new additions to taxable income in the year), but only when the result is greater than zero. Row U calculates whether there are any losses remaining to carry forward to the following year by adding together Rows R and S and subtracting Row T. Note: numbers in brackets are negative and those without brackets are positive. V Tax to pay

44

102

80

2

30

25

8

Let’s say the tax rate is 40%, so, in each year, the tax to pay is 40% of that year’s taxable income (Row T). W Contractor’s revenue entitlement

134

400

548 470 216 204 159

97

45

The sum of the contractor’s cost recovery (Row L), contractor’s share of Excess Cost Oil (Row M), Contractor’s share of total Profit Oil (Row N) and contractor’s tax to pay (Row V). The Contractor’s revenue entitlement includes entitlement to production relating to tax that, although he doesn’t pay, he is nevertheless liable for. That production can usually be reported in the associated companies’ (that make up the contractor) reports and accounts as theirs. Contractor’s production entitlement in millions of barrels

1.49 4.45 6.09 5.22 2.40 2.26 1.77 1.07

0.50

Contractor’s revenue entitlement (Row W) divided by oil price (Row A) equals the Contractor’s entitlement to oil. Net cash flow X Revenue (from Row W) Exploration and appraisal costs (from Row C) Development capital costs (from Row D) Production operating costs (from Row E)

134

400

548 470 216 204 159

97

45

(50)

(50)

(30) (250) (300) (150) (50) (50) (50) (50) (50) (50) (50)

(100)

Decommissioning costs (from Row F) Y Tax (from Row V) Net cash flow

(44) (102) (80) (30) (250) (216) 156

(2)

(30) (25)

396 340 165 124

84

(8) 39

(5)

(100)

Note in this table that amounts are designated positive or negative (denoted by brackets) depending on whether they represent positive cash flow (into the company) or negative cash flow (out of the company). So in Years 1, 2, 3 and 12 there’s net negative cash flow (out of the company) and in all the other years there’s net positive cash flow (into it). Also note that the only part tax plays is to increase revenue (Row X) by the amount that tax represents, and that amount is then subtracted again in Row Y. Appendix 4.2.3 – Production sharing contract: a cost recovery ceiling exists, contractor is liable for decommissioning costs, and tax is paid on behalf of the contractor by the government using a profit oil gross-up calculation. Year Year Year Year Year Year Year Year Year Year

Year

Year

1

2

3

4

5

6

7

8

9

10

11

3

8

10

10

8

6

4

2

1

A Oil price, USD per barrel Price the oil is sold for.

90

90

90

90

90

90

90

90

90

B Revenue generated, USD million

270

720 900 900 720 540 360

180

90

Production, millions of barrels of oil Oil production from the oil field.

12

Production multiplied by oil price; so USD 270 million in Year 3 is 3 million barrels of oil x USD 90 per barrel. C Exploration and appraisal costs, USD million

30

Exploration and appraisal costs associated with the oil field. D Development capital costs, USD million

250

300

150

50

50

The costs incurred to develop it. E Production operating costs, USD million

50

50

50

50

50

50

50

All the costs associated with producing and transporting the oil to where it’s sold. F Decommissioning costs, USD million

100

The cost of decommissioning the oil field’s wells and production facilities G Royalty, USD million

27

72

90

90

72

54

36

18

9

Let’s say there’s a royalty which is 10%, so 10% of revenue (Row B) Production Sharing Contract calculations, all USD million unless otherwise stated: H Production available for Contractor’s cost recovery and profit

243

648 810 810 648 486 324

162

81

Royalty is subtracted from the oil field’s revenue before the production sharing contract’s cost oil and profit oil calculations; so revenue generated (Row B) minus royalty (Row G). Although this is ‘production’ available for the contractor’s cost recovery and profit, it’s measured here in USD million. In order to do these calculations, either production has to be converted into money (as is done here) or else money has to be converted into production (the Contractor spends money, not barrels of oil) by dividing it by the oil price (so, for example, USD 90 million equates to 1 million barrels of oil if the oil price is USD 90/bbl). The calculations can be done in either ‘currency’. I Maximum production available for Contractor’s cost recovery

97

259 324 324 259 194 130

65

32

Let’s say that only 40% of production is made available for the purpose of Contractor’s cost recovery, in other words there’s a cost recovery ceiling of 40%, so 40% of Row H. Cost oil: Development capital additions to the cost pool from:

Development capital costs incurred in Year 2

250

Development capital costs incurred in Year 3

300 150

Development capital costs incurred in Year 4 J Total development capital additions to the cost pool:

550

150

Let’s say development capital costs are added to the cost pool only when the associated assets are placed into service; and let’s say the assets associated with the capital spent in Years 2 and 3 become operational when production begins and the assets associated with the capital spent in Year 4 become operational immediately. Note that, in this example, these costs join the cost pool as a single amount and aren’t drip-fed into it. K Decommissioning cost additions to the cost pool

20

20

20

20

20

Decommissioning costs aren’t always explicitly mentioned in production sharing contracts and, if they aren’t, decommissioning is assumed to be the responsibility of governments because, usually, assets (e.g., wells, production facilities, pipelines) become their property either immediately or once the contractor has recovered its costs of them. Here, let’s say the contract stipulates that decommissioning costs are the contractor’s responsibility and they’re added to the cost pool in five equal amounts over the last 5 years of production from the oil field (if the contract does specify that decommissioning is the contractor’s responsibility, it will explain how they’re recovered). Opening balance of cost pool (see Note 1 below)

30

533 474 200

Development capital costs (from Row J)

550

150

Production operating costs (from Row E)

50

50

Additions to the cost pool from: Exploration and appraisal costs (from Row C)

30

30

L Contractor’s cost recovery (see Note 2 below) Closing balance of cost pool (see Note 3 below)

43

30

50

50

Decommissioning costs (from Row K) Total additions to the cost pool

5

30

30

50

50

50

50

50

50

20

20

20

20

20

600

200

50

70

70

70

70

70

97

259 324 250

70

70

70

65

32

533

474 200

5

43

43

Note 1: The opening balance of the cost pool in any year is its closing balance in the previous year. It’s automatically zero in the contract’s first year. Note 2: The contractor’s cost recovery is the lesser of a) the maximum production available for the contractor’s cost recovery (Row I) and b) the cost pool’s balance, which is the sum of its opening balance plus the total additions for the year (in other words all the costs the contractor has spent but hasn’t recovered yet). If the maximum production available for the contractor’s cost recovery is less than what’s in the cost pool, then cost recovery is limited to that amount. If, however, the maximum available production is more than what’s in the cost pool, then all of it can be recovered. A useful check to confirm the integrity of these calculations is to compare the total costs incurred and total cost

recovery, which ought to be the same. Here, however, the contractor has spent USD 1.280 million but only recovers USD 1.237 million because some costs (late in the oil field’s life) aren’t being fully recovered. Note 3: The closing balance of the cost pool in any year is the opening balance for that year, plus the total additions for the year, less the amount that has been recovered by the contractor. Excess Cost Oil

74

189 124

60

Because, here, a certain proportion of the production available for contractor’s cost recovery and profit is being allocated to contractor’s cost recovery, there will be periods when that’s too much, and then the question is what does the contract do with the surplus. Usually what it does is treat it as if it was profit oil —divided between the contractor and the government—sometimes in exactly the same way profit oil is divided, although sometimes differently. M Contractor’s share of Excess Cost Oil

19

47

31

15

Let’s assume that excess cost oil is treated in exactly the same way as profit oil (and the contractor receives 25% of it). Profit oil: Total Profit Oil

146

389 486 486 389 292 194

97

49

If 40% of the production available for contractor’s cost recovery and profit has been allocated to contractor’s cost recovery, then 60% is available for profit oil, so 60% of Row H. N Contractor’s share of total profit oil

36

97

122 122

97

73

49

24

12

Let’s assume the contractor receives 25% of the total profit oil (although, as said in Chapter 4, the mechanism can be much more complicated than that). Tax calculations: O Tax to pay

24

65

81

93

96

69

42

16

8

The calculation here is the contractor’s overall profit, which is his shares of excess Cost Oil and total profit Oil (Row M + Row N) × [tax rate / (1 – tax rate)]; and let’s say the tax rate is 40%. So in Year 3: 36 × [40% / (1 − 40%)] which equals 24 (we can ignore Row M - contractor’s share of excess cost oil because it’s zero in that year). The calculation can be checked back again by adding Rows M, N, and O together, which represents the pretax equivalent of the contractor’s profit (in Year 3 that would be 36 + 24 equals USD 60 million), then take 60% of that amount, which is the amount of pretax profit oil that the contractor gets to keep net of tax, so 60% of USD 60 million, which is USD 36 million (the same as Year 3’s contractor’s share of total profit oil). P Contractor’s revenue entitlement

158

421 527 483 311 243 176

105

53

The sum of the contractor’s cost recovery (Row L), contractor’s share of excess cost oil (Row M), contractor’s share of total profit oil (Row N) and the contractor’s tax to pay (Row O). The contractor’s revenue entitlement includes entitlement to production relating to tax that, although he doesn’t pay, he is nevertheless liable for. That production can usually be reported in the associated companies’ (that make up the contractor) reports and accounts as theirs. Contractor’s production entitlement in millions of barrels

1.76 4.68 5.85 5.37 3.45 2.70 1.95 1.17

0.59

Contractor’s revenue entitlement (Row P) divided by oil price (Row A) equals the Contractor’s entitlement to oil. Net cash flow Q Revenue (from Row P) Exploration and appraisal costs

158 (30)

421 527 483 311 243 176

105

53

(from Row C) Development capital costs (from Row D)

(250) (300) (150)

Production operating costs (from Row E)

(50) (50) (50) (50) (50) (50) (50)

(50)

(50)

Decommissioning costs (from Row F)

(100) (24) (65) (81) (93) (96) (69) (42)

R Tax (from Row O) Net cash flow

(30) (250) (216) 156 396 340 165 124

84

(16)

(8)

39

(5)

(100)

Note in this table that amounts are designated positive or negative (denoted by brackets) depending on whether they represent positive cash flow (into the company) or negative cash flow (out of the company). So in Years 1, 2, 3 and 12 there’s net negative cash flow (out of the company) and in all the other years there’s net positive cash flow (into it). Also note that the only part tax plays is to increase revenue (Row Q) by the amount that tax represents, and that amount is then subtracted again in Row R.

Appendix 14.1–Exploration This appendix works through Chapter 14’s final example, which, as a reminder, was this.

We have an exploration target for which our geoscientists have identified an exploration well location and, potentially, three appraisal well locations. And we got as far, in Chapter 14, as constructing a decision tree that looked like this.

It isn’t populated and is the result of only the left-to-right pass that maps out the exploration target’s possible futures. So now it needs to be populated and the expected monetary value of each decision calculated to tell us whether we should drill the exploration well or not, and what, if any, appraisal wells we should drill in the event of a discovery. Step 1: Would We Develop and Produce It? We start at the tree’s right-hand side and first of all calculate how much richer or poorer developing and producing each possible realisation of the field, in the case of a discovery, would make us.

• So the net present values are the sums of the discounted cash flows associated with the field’s development and production only, discounted to when the decision to develop and produce it is taken (which may be up to a few years after the exploration well is drilled). This establishes the field realisations that would proceed, when the time comes, and would make the company richer, and those that wouldn’t because doing so would make it poorer. Net present values that are negative can therefore be replaced with zeros because they wouldn’t proceed. • Different realisations with the same field sizes don’t necessarily have the same net present values because uncertainty may dictate that the company proceeds differently and doesn’t, for example, build a dedicated (and expensive)

production platform if there still remains a real possibility that too little oil has been found to justify it. This can be seen at the top of the tree where a onebillion-barrel realisation has a net present value of USD 10 billion, whereas very close to the bottom, another 1-billion-barrel realisation has a much lower net present value (USD 3 billion). The difference is we’re confident that in the former—after three successful appraisal wells—the field can’t realistically hold less than 700 million barrels and so we’re sure it justifies that dedicated (and expensive) platform; but there’s every chance it holds less than that in the latter, and so if we proceed to development and production at such an early stage with so little understanding of the field (after the discovery but without any appraisal), we’d be bound to do so much more cautiously. Step 2: Should We Drill the Third Appraisal Well? We continue from right to left, and the next decision we come to is whether it makes sense to drill a third appraisal well. But we need more information before we can answer that; we need to know the likelihood that the third appraisal well will find oil, taking into account whether previous appraisal wells have or haven’t found any, and we also need the values of each possible realisation relating to the new decision in hand; which I’ve inserted here:

• Branches that have no relevance to this decision are denoted as not applicable at their end-points (n/a). • The net present values now are the sums of the discounted cash flows associated with the field’s possible futures forward of the third appraisal well’s decision, discounted to when that decision is taken. So the cash flows associated with development and production are discounted further back in time than they were before, and the “Yes, we drill it” branches include the cash flows relating to the appraisal well itself. Hence, they’re different from the previous step’s equivalent net present values. • Out of sheer laziness I’ve assumed the likelihood that the third appraisal well will find oil is 40% and that it won’t find any is 60%, whatever the results of previous appraisal wells (although the real assumptions would be sourced from our geoscientists). • The expected monetary values on the far right of the tree are the net present values multiplied by the likelihood of the realisations occurring in the context of

this decision only. So, for example, at the very top of the tree:

• Now we can calculate two total expected monetary values for each potential third appraisal well decision—one associated with “Yes, we drill it,” and the other associated with “No, we don’t,” which are the red numbers above and below the four third appraisal well’s decision squares. The numbers above the squares are the expected monetary values of the Yes branches and are the sums of the individual expected monetary values for all the possible realisations leading from them, and the numbers below the squares are the expected monetary values of the No branches and are the sums of the individual expected monetary values resulting from those choices. So for the uppermost decision of the third appraisal well, for example: USD 6,879 million = USD 1,087 million + [down to] + USD 976 million (the first six individual expected monetary values from the top). USD 7,091 million = USD 2,727 million + [down to] + USD 1,636 million (the next three individual expected monetary values). • In every case, the numbers below the squares are greater than those above them, although in the third scenario down (representing the first appraisal well not finding oil but the second one doing so), the numbers are close. We therefore conclude that there are no circumstances in which drilling the third appraisal well makes sense, whatever the result of previous wells, because the expected monetary value of the field is greater if we don’t. Step 3: Should We Drill the Second Appraisal Well? Continuing from right to left, we must next decide whether to drill the second appraisal well. Once again, however, we need more information before we can do that. We need to know the likelihood of the second appraisal well finding oil, taking

into account whether the first appraisal well found any, and we also need the values of each possible realisation relating to the new decision in hand, which I’ve entered below.

• The net present values now are the sums of the discounted cash flows associated with the field’s possible futures forward of the second appraisal well’s decision, discounted to when that decision is taken. So all the cash flows associated with development and production are discounted even further back in time than before, any cash flows associated with the third appraisal well are discounted (which are none here because we won’t drill it), and the “Yes, we drill it” branches of the second appraisal well include the cash flows relating to that well. Hence, they’re different again from the previous step’s equivalent net present values. • The expected monetary values relating to the third appraisal well remain unchanged. Those values are still the right ones because that decision is made when it’s made, not right now.

• Still being lazy, I assume that the likelihood of the second appraisal well finding oil is 40% and of it not finding any is 60%, whatever the result of the first appraisal well. • The expected monetary values on the far right of the tree are the net present values multiplied by the likelihood of the realisations occurring in the context of this decision only. They show as zero wherever branches are redundant—in other words, not chosen (every realisation relating to drilling the third appraisal well shows as zero, therefore, because we won’t drill it). So, for example, the first non-zero value from the top—the seventh down— is:

• Now we can calculate two total expected monetary values for each potential second appraisal well decision—one associated with “Yes, we drill it,” and the other associated with “No, we don’t,” which are the red numbers above and below the second appraisal well’s two decision squares. The numbers above the squares are the expected monetary values of the Yes branches and are the sums of the individual expected monetary values for all the possible realisations leading from them, and the numbers below the squares are the expected monetary values of the No branches and are the sums of the individual expected monetary values resulting from those choices. So for the second appraisal well’s uppermost decision, for example, we have: USD 4,314 million = USD 0 + [down to] + USD 292 million (the first 18 individual expected monetary values from the top). USD 3,537 million = USD 1,983 million + [down to] + USD 496 million (the next three individual expected monetary values). • In both cases, the numbers above the squares are greater than those below them, although they’re close in the event the first appraisal well doesn’t find oil. So we conclude that we should drill the second appraisal well, whatever the

result of the first one.

Step 4: Should We Drill the First Appraisal Well? Carrying on, we must next decide whether to drill the first appraisal well. As before, more information is needed, however. We need to know the likelihood of the first appraisal well finding oil and also the values of each possible realisation relating to the new decision in hand, which are shown here:

• The net present values now are the sums of the discounted cash flows associated with the field’s possible futures forward of the first appraisal well’s decision, discounted to when that decision is taken. So all the cash flows associated with development and production are discounted further back in time than before, any cash flows associated with the second and third appraisal wells are discounted (which are none for the third appraisal well because we won’t drill it), and the “Yes, we drill it” branch of the first appraisal well includes the cash flows relating to that well. Hence, they’re different again from the

previous, equivalent net present values. • The expected monetary values relating to the second and third appraisal wells remain unchanged. These values are still the right ones because those decisions are made when they’re made, not right now. • Lazy to the last, I assume the likelihood of the first appraisal well finding oil is 40% and of it not finding any is 60%. • The expected monetary values on the far right of the tree are the net present values multiplied by the likelihood of the realisations occurring in the context of this decision only. As before, however, they show as zero where branches are redundant—in other words, not chosen. So, for example, the first non-zero value from the top—the seventh down— is:

• Now we can calculate two total expected monetary values for the first appraisal well decision—one associated with “Yes, we drill it,” and the other associated with “No, we don’t,” which are the red numbers above and below the first appraisal well’s decision square. The number above it is the expected monetary value of the Yes branch and is the sum of the individual expected monetary values for all the possible realisations leading from it, and the number below is the expected monetary values of the No branch and is the sum of the individual expected monetary values resulting from that choice. These values are: USD 1,951 million = the sum of all the individual expected monetary values from the top, except for the last three. USD 947 million = USD 676 million + [down to] + USD 0 (the last three individual expected monetary values). • The upper expected monetary value is greater, and so we conclude that we should drill the first appraisal well.

Step 5: Should We Drill the Exploration Well? Lastly, we reach the left-hand side of the tree, and the final decision to make is whether to drill the exploration well. Once again, though, more information is needed. We need to know the likelihood of the exploration well finding oil and the values of each possible realisation relating to the new decision in hand, which are shown here:

• The net present values now are the sums of the discounted cash flows associated with the field’s possible futures forward of the exploration well’s decision, discounted to when that decision is taken. So all the cash flows associated with development and production are discounted yet further back in time, any cash flows associated with appraisal wells are discounted, and the “Yes, we drill it” branch of the exploration well includes the cash flows relating to that well. Hence, they’re different again from previous, equivalent net present values. • The expected monetary values relating to the appraisal wells remain unchanged because those decisions are made when they’re made, not right now. • I assume the likelihood of the exploration well finding oil is 25% and it not finding

any is 75%. • The expected monetary values on the far right of the tree are the net present values multiplied by the likelihood of the realisations occurring in the context of this decision only. As before, however, they show as zero where branches are redundant—in other words, not chosen. • So, for example, the first non-zero value from the top—the seventh down—is:

• Now we can evaluate the two total expected monetary values for the exploration well decision—one associated with “Yes, we drill it,” and the other associated with “No, we don’t,” which are the red numbers above and below the exploration well’s decision square. The number above it is the expected monetary value of the Yes branch and is the sum of the individual expected monetary values for all the possible realisations leading from it, and the number below is the expected monetary values of the No branch. These values are: USD 414 million = the sum of all the individual expected monetary values from the top, except for the very last one (which is a zero). USD 0 = the expected monetary value of doing nothing, “Don’t drill the well.” • The upper expected monetary value is greater, and so we conclude that we should drill the exploration well. As I noted in Chapter 14, software thankfully exists that makes the process shown in this appendix rather less laborious. Finally, given that we now know how we should proceed regarding exploration and appraisal here, we could, if we wanted to, work out what minimum field size would make us richer by drawing a graph of the net present values from step 1—but only those at the end of the branches relating to drilling the exploration well and the first two appraisal wells—(on the y-axis) against their field sizes (on the x-axis). The point where the relationship crosses the x-axis is that minimum size. We could also, if we wanted to, haul our geoscientists out of the pub and ask them what the likelihood is of

us not only finding oil, but finding that much or more—and that’s the (potential) field’s POSc, commercial probability of success.

Appendix 15.1–Sensitivity Analysis This appendix is dedicated to probabilistic petroleum economics and draws from my own experiences of using it in anger (by which I mean for real, not that using it makes me angry) many times. Probabilistic petroleum economics is different. There really are no hard and fast rules; there’s no textbook approach. And that’s why I’d much rather consider this appendix is an indication of how to go about it rather than it representing a set of rules. A worthy substitute for the absence of rules, however, is the constant reminder I give myself of what I’m trying to do, which is help decision makers make the right decisions. It isn’t an academic exercise that offers an opportunity to gain a PhD. Nor is it a tick-in-the-box exercise: “The board wants to see an S-curve [the cumulative frequency distribution in Chapter 15 that looks rather like a letter S], so we’d better make one up.” I’ve seen the task approached both ways, and neither is helpful. Rather, the task is a practical exercise to capture the uncertainty surrounding the assumptions that are made in calculating a net present value, say, as comprehensively but realistically as possible—comprehensively because of course we want to capture all the uncertainty there is, but realistically because it has to be done in such a way that doesn’t take an age and an army of people to do, and also that’s understandable, especially to decision makers.

My General Process I start with a list of all the forecasts, predictions, estimates, and other assumptions we’ve made. Then I create a tornado chart, although one that isn’t necessarily especially refined. I talk to reservoir engineers about production, to cost engineers about capital costs, to operations teams about operating costs, to the commercial group about transportation costs, to asset managers about further growth we may be able to exploit if that’s relevant, to the tax department if needs be—to anybody, in fact, who might say, “But it may not turn out like that, and if it doesn’t…”, then they stop and pull a face as if to say anything could happen. The tornado chart is to find out what matters and what doesn’t, because I want to concentrate on the former and not spend long on the latter. Let’s say the conclusion—for our purpose here, anyway—is that quite a few things seem to matter. The best example I can remember where that was certainly the case was a deal that involved the company I worked for swapping some of its oil fields for some of another company’s. As well as there being plenty of uncertainties associated

with the individual fields involved in the swap, the oil price—which one might have thought was a common factor for all of them—also turned out to have a material impact because the outgoing fields produced relatively few near-term, high (post-tax) value barrels, whereas the incoming fields produced relatively many long-term, low (post-tax) value barrels; so, for example, the possibility of near-term oil price spikes was bad news. Next, the material uncertainties that the net present value (say) calculations will take into account have to be described, which I do as follows. 1. I use individual project scenarios to capture production-related uncertainty. I prefer doing that over modelling production profiles as a continuous range (where any profile can occur between two extremes) because it avoids having to include logic in the model to control “structural” decisions such as the number of wells that are required or how a discovery would be developed (an optimistic field size may be developed differently than a pessimistic one). 2. I can then apply probabilities of those scenarios occurring. If there are three scenarios and we believe those probabilities are 30%, 40%, and 30%, for example, my (Excel) model would have a cell that returns 1, 2, or 3 during the Monte Carlo simulation with those probabilities and that cell controls the scenario the model selects to use each time it calculates a result. 3. Within each scenario, I then capture the uncertainty relating to other forecasts, predictions, estimates, and assumptions as being either continuous in nature (able take any value within reason) or discrete (can be one of only a few possibilities). Continuous Forecasts include, for example, oil price, capital and operating costs that can take any value within a considered range, and numerous forms exist to capture the relative likelihood of values occurring within that range (based on historical data or someone’s judgment). Often the form takes the shape of an expectation about which there’s a high degree of indifference immediately either side of it, with the likelihood of values higher or lower than that tailing away to relative insignificance, which is why normal and lognormal distributions are favourites to use.

I must say, however, that I often end up using a custom-designed form, just

because they represent the reality of our understanding of things better. Oil price tends to be a victim of that, whose form might, for me, look something like this:

Here, there’s an extended region of indifference around what we think the oil price will be next year (in the left-hand form); and in a few years’ time, that region is extended even further (in the right-hand form). There simply isn’t a single price that we have more faith in than any other. At the very best, all we can say is we think it’ll be somewhere within this band, and then the likelihood of prices higher or lower than that tails off outside it. Furthermore, the form relating to near-term years ahead might be asymmetrical if we think the oil price could be much higher than lower for some reason over that period; however, the long-term picture may be more symmetrical if we have to admit that we have no idea what the future holds beyond the next few years. Discrete Forecasts work in much the same way as the earlier production-related uncertainty. In the case, say, of the two possible branch remittance tax rates from Chapter 15 that might apply, I’d have a cell that returns a 1 or 2 in proportion to the probability that we think each rate will apply. That cell would then control the selection of one or other rate in the calculations during the Monte Carlo simulation.

Appendix 17.1—More Difficult Lease or Purchase Evaluations The example of a lease or purchase evaluation in Chapter 17 was a simple one, and you were probably thinking throughout it, “Yes, but what if the lease includes maintenance and operating costs?” Or, “What if my company has a different tax position, if it isn’t tax paying and so won’t immediately benefit from tax deductions?” Or, “What if I’m in a production sharing contract world rather than a tax/royalty one?” And probably more to boot. Well, then it becomes harder and eventually impossible to do the calculations as they were straightforwardly written out there, and we have to use our cash flow model to help us isolate the financing cash flows. Four model scenarios need to be run: two with the model set up to evaluate the lease and two with it set up to evaluate the purchase. For the latter, the model may

need reconfiguring to accommodate the oil company’s loan, interest, and repayment of the loan, although it isn’t uncommon that the loan is taken out in a different country to that in which the oil field resides and therefore is governed by different fiscal terms, in which case those calculations have to be dealt with separately, outside the cash flow model. The two lease scenarios are: • With the lease costs included • With the lease costs excluded Then the difference between the two resulting net cash flows is the cash flow that relates solely to obtaining the asset by leasing it, and is the equivalent of Line 3 in Chapter 17’s example. The two purchase scenarios are • With all the purchase-related costs included • With all the purchase-related costs excluded Whatever elements are included and then excluded in the lease scenarios have to be exactly mirrored in the purchase scenarios by including and then excluding their equivalent cash flows. So, for example, if the lease includes operating and maintenance, those same equivalent costs must be included and excluded, respectively, in the two purchase scenarios. Then, the difference between the two resulting net cash flows is the cash flow that relates solely to obtaining the asset by purchasing it and is the equivalent of Line 12 in the simple example in Chapter 17. Here’s another example (and in this case the loan is dealt with outside the cashflow model). Year Year Year Year Year Year Year Year Year Year Year Year Year 1 2 3 4 5 6 7 8 9 10 11 12 13 1 Lease, all cash flows included

(300) (400) (120) 593

223

128

109

38

24

15

8

(3)

2 Lease, excluding cash flows particular to lease

(300) (400)

15

549

191

148

130

58

44

36

28

23

3 Lease financing costs, undiscounted

(135)

44

32

(20)

(20)

(20)

(20)

(20)

(20)

(25)

4 Lease financing (176) costs, discounted

(128)

40

29

(18)

(17)

(17)

(16)

(16)

(15)

(19)

(550) (750) (185) 689

471

235

324

51

37

28

21

15

5 Purchase, all cash flows

11

included 6 Purchase, excluding cash flows particular to purchase

(300) (400)

15

549

191

148

130

58

44

36

28

23

18

7 Purchase financing costs, undiscounted

(250) (350) (200) 140

281

87

194

(7)

(7)

(7)

(7)

(7)

(7)

Note: Lines 5, 6 and 7 exclude the loan 8 Borrow money to build FPSO

250

350

150

9 Interest on loan

(10)

(24)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

(30)

10 Tax savings from interest

3

7

9

9

9

9

9

9

9

9

9

9

9

11 Repay loan

(750)

12 Total purchase financing costs, undiscounted 13 Total purchase financing costs, discounted 14 Discount factors

(215)

(7)

(17)

(71)

119

260

66

173

(28)

(28)

(28)

(28)

(28) (778)

(7)

(16)

(67)

110

233

57

147

(23)

(23)

(22)

(22)

(21) (559)

1.000 1.028 1.057 1.086 1.117 1.148 1.180 1.213 1.247 1.282 1.318 1.355 1.393

I don’t propose to say much about this situation’s background or why the cash flows look like they do; that isn’t the point. Suffice it to say that we’re faced with an FPSO (reminder, a floating production, storage, and offloading vessel) lease or purchase decision for a potential oil field development in Indonesia, where production sharing contracts prevail. We are a company based in the US, however, and any loan (in the purchase case) will be taken out there. The lease in this case requires the lessor to maintain and operate the vessel as well as provide it. Lines 1 to 4 in the table concern the lease. Line 1 is the project’s net cash flow (derived from our cash flow model), including all the relevant cash flows relating to, and resulting from, the lease; so it includes the lease costs, their cost recovery, any tax benefits the lease costs provide, and so on. Line 2 is the same, but excludes the lease costs. The difference between Lines 1 and 2 is therefore the overall impact of the lease on the project’s total net cash flow (Line 3). Lines 5 to 13 concern the purchase. Lines 5 to 7 repeat the process just described

in the context of the lease, but for the purchase. Line 5 is the project’s net cash flow, including all the relevant cash flows relating to, and resulting from, the purchase; so it includes the capital and operating costs of building, maintaining, and operating the FPSO, their cost recovery, any tax benefits that those costs provide, and so on. Line 6 is the same cash flow but excludes the FPSO’s capital and operating costs. The difference between Lines 5 and 6 is therefore the overall impact the purchase has on the project’s total net cash flow (Line 7). Lines 8 to 11 concern the loan, which works, in principle, as before but whose interest and tax savings are evaluated outside the cash flow model as a result of the company being based in the US and not Indonesia. Line 12 is the project’s net cash flow in the purchase case (Line 7) plus the loanrelated cash flows (lines 8 to 11). All that leaves us to do, then, is discount Lines 3 and 12 exactly as before, by using a discount rate based on borrowing in the US, after tax, because that’s where the borrowing is being done. And our conclusion, having done that, is that we should lease the FPSO because doing so has a lower cost—a smaller negative net present value—than purchasing it (USD 176 million vs. USD 215 million).

INDEX A acquisitions and divestments acquisition cost, 113–114 assets and liabilities, 116 completion statement, 115 data room, 114, 115 disposal proceeds, 113–114 flyer, 114 information memorandum, 114 legal papers, 115 net present values, 115, 116 questions and answers (Q&A), 115 taxable losses, 116

B beginning-year discounting, 59 branch office, 29 branch remittance taxes, 30 brand-new cash flows, 6

C capital asset pricing model (CAPM), 40–41 capital costs, 20, 21 cash flow between bank accounts, 6–7 brand-new cash flows, 6 company ownership structure and internal cash flow, 1, 2 definition, 1 fiscal terms, 5 investigation, 2 relevant cash flow (see relevant cash flow) significance, 2 cash tax, 22 clean cash flows, 49 concession fiscal terms, 37, 38 cash tax, 22 vs. contractual, 16–17 current tax liability, 22 decommissioning, 21, 22 definition, 16 production taxes, 18

profit-based taxes deductions, 19–22 depreciation, 20 feature, 18 value, 19 ring-fence consolidation, 23, 143–145 definition, 22–23 individual oil field, 23, 139–142 royalty, 17–18 consolidation, 23 contractual fiscal terms vs. concession, 16–17 definition, 16 PSC bonuses, 27 cost oil, 25, 26 cost pools, 25, 26 cost recovery, 25 cost recovery ceiling, 26 decommissioning costs, 25–26 definition, 24 entitlements, 26, 28 profit oil, 25–27 PSC return, 24–25 tax calculation, 27 service contracts, 28–29 corporate costs, 51 cost of capital, 52 cost oil, 25, 26 cost pools, 25, 26 cost recovery, 25 cost recovery ceiling, 26 current tax liability, 22

D decision trees, 95–100, 102 decommissioning costs, 21, 22, 25–26, 38 depreciation reducing-balance depreciation, 20 straight-line depreciation, 20 discounting, 42–43, 137 arithmetic of, 58–60 beginning-year discounting, 59 calculations, 52 discount date, 58 discount rates, 52, 56, 57

end-year discounting, 59 hurdle rate, 52, 53, 55, 56 measures, 67–68 mid-year discounting, 59–60 net present value, 53, 54 positive cash flow, 54, 55 present value, 53, 54 and undiscounted net cash flows, 55–57 WACC, 58 discount rates, 52, 56, 57 divestments. See acquisitions and divestments double tax treaties (DTTs), 30–31

E end-year discounting, 59 escalation, 69 excess cost oil, 26 exemption regime, 32 exploration costs, 51 exploration well appraisal well expected monetary value, 97–100, 169 first appraisal well, 167–168 second appraisal well, 165–166 third appraisal well, 163–165 arithmetic, 90–91 cost deduction, 89 cost pools, 90 decision trees, 95–100, 102 development and production, 92 expected monetary value, 99–101 exploration target, 97, 101, 102 geological vs. commercial likelihood of success, 93–95 geologists’ revised probability, 91, 92 geoscientists’ estimation, 90, 91 high-case estimate, 92 low-case estimate, 92 mid-case estimate, 92 net present values, 90, 100, 169 tax, 89

F financing costs, 123–126 financing decision, 121, 122, 126 fiscal terms concession (see concession fiscal terms) contractual (see contractual fiscal terms)

definition, 5 foreign exchange conversion rates, 73–74 foreign tax credit regime, 32

H hurdle rate additional return on projects, 71–72 corporate-based hurdle rate, 56 definition, 52 WACC, 71

I incremental economics definition, 82 new discovery, development and production consolidated position, 85–86 post-tax cash flow, undiscounted and discounted, 86 rest of the company, 84–85 separately, on its own, 83–84 process, 82 rules of thumb, 82 indirect taxes, 33 inflation vs. escalation, 69 hurdle rate additional return on projects, 71–72 WACC, 71 real and nominal cash flow difference, 70 information collection dealing with people, 131–132 discussions, 132 sixty-four-thousand-dollar question, 132–133 internal rate of return (IRR), 64–65 investment decision, 121, 122

L lease borrowing costs, 123 capital element, 120–121 definition, 119 evaluations, 174–176 financing costs, 123–126 financing decision, 121, 122, 126 investment decision, 121, 122 operating element, 120–121 purpose, 119–120

tax deductions, 120 tax depreciation, 120 treasury team, 126

M mid-year discounting, 59–60 money sources, 39 Monte Carlo simulation, 110

N native currency, 73 net cash flow, 37–38 non-cash cost, 21 non-professional petroleum economists, 135

O operating costs, 20, 21 opportunity cost, 8 opportunity cost of capital, 52

P payback, 63–64, 67 present value index (PVI), 65–67 probabilistic petroleum economics cumulative frequency distribution, net present values, 109 Monte Carlo simulation, 110 probability density function, net present values, 108, 109 production sharing contracts (PSC) bonuses, 27 cost oil, 25, 26 cost pools, 25, 26, 146–150 cost recovery, 25 cost recovery ceiling, 26, 151–160 decommissioning costs, 25–26 definition, 24 entitlements, 26, 28 expected monetary value, 170 profit oil, 25–27 profit oil gross-up calculation, 157–160 PSC return, 24–25 tax calculation, 27 profit-based taxes deductions, 19–22 depreciation, 20 feature, 18

value, 19 profit oil, 25–27 project costs, 51–52 PSC. See production sharing contracts (PSC) purchase decisions, 122–126, 174–176 PVI. See present value index (PVI)

Q qualitative models, 75–77, 137 quantitative models, 77–78

R real vs. nominal cash flow, 70 reducing-balance depreciation, 20 relevant cash flow between bank accounts, 6–7 brand-new cash flows, 6 difference principle, 48 directors, 48 new field development, 7–8 office space, 48 in oil field appraisal, 11, 12 decommissioning, 11, 14 development, 11, 13 exploration, 11, 12 production, 11, 13 opportunity cost, 8 project team, 48 size and timing, 9 tax, 8 reports and presentations, 129–130 return on investment, 39–42 ring-fence consolidation, 23, 143–145 definition, 22–23 individual oil field, 23, 139–142 royalty, 17–18

S sensitivity analysis continuous forecasts, 172 discrete forecasts, 173 graph, 105–106 net present value calculations, 171–172 probabilistic petroleum economics, 108–111

spider diagrams, 106–107 tornado charts, 107–108, 171 service contracts, 28–29 spider diagrams, 106–107 straight-line depreciation, 20 subsidiary, 29 Swanson’s rule, 94, 95

T tax, 8 branch remittance taxes, 30 exemption regime, 32 foreign tax credit regime, 32 indirect taxes, 33 tax treaties, 30–31 withholding taxes, 30, 32 tax/royalty. See concession fiscal terms tornado charts, 107–108, 171

U unclean cash flows, 49 unit costs, 63, 67

W weighted average cost of capital (WACC), 42, 52, 58, 71, 126 withholding taxes, 30, 43, 32