1. The Yi Ching
Those familiar with the ancient Yi Ching, also called the Book of Changes, are sometimes taken to divine a person, an entity, or the outcome of a process, by contemplating the issue, then recording six tosses of three coins, to obtain the appropriate hexagram. The basis for this divination is the Taoist belief that all elements of the universe are fully expressed, and connected, in the Tao. So it is logical to ask for answers to difficult questions. I was contemplating Exxon Mobil (NYSE:XOM), as a genuine puzzle of 2016. What is this company like? And, where is it headed?
If you try this at home as I did, some of you may arrive at Phi, the twelfth hexagram in a set of 64 possible permutations. Its interpretation is not favorable, as the name might indicate: Standstill, or Stagnation. In a nutshell (from James Legge's old book), "The great gone, and the little come".
The hexagrams are interpreted as a whole, using the bottom and the top trigrams, and then line by line. A "six" is a broken line; a "nine" is a solid line. An interpretation of the first (bottom) broken line, by Wilhelm, 1950, bears an eerie relevance to the current Exxon Mobil situation, in the last 2 sentences.
Namely, Chairman and CEO Rex W. Tillerson is retiring, and Darren W. Wood will likely succeed, the first non-upstream person to lead Exxon Mobil.
Leaving this exciting track for the moment, we restate our two questions. What is XOM like? And, where is it headed? For answers, we take a look at performance, as provided by the recent XOM 10 K SEC filings since 2010. By performance we mean proved reserves, production, revenues, and financial results.
For any of the large oil companies, there are three sectors of activity: upstream or exploration and production, downstream or refining to finished product, and synthetic petrochemical. Without question, until recently, upstream was king of earning. And within upstream, crude production, by far, was ruler. For a long time, crude oil was a high value product coming to the surface on its own, easily flowing, and cheaply transportable all over the world. Those times are gone.
2. Proved Reserves
As reported in the 10 K filings, total proved reserves are the sum of developed and undeveloped reserves. For Exxon Mobil, the total reserves in 5 product categories are plotted in F.1. Over the six year period preceding, the total reserves are remarkably flat, which shows a concern to keep the totals as constant as possible. This gives the company the feel of sailing at even keel, keeping its stores replenished, while producing significant volumes.
So, total of all reserves is a flat line, about 25 billion barrels of oil equivalent (BOE) including crude, natural gas liquids (NGL), Bitumen from oil sands, synthetic oil, and natural gas (NG). The composition of the total mix varies with time, in a rather systematic manner. In general, bitumen is gradually rising, while NG reserves are declining.
The tranquility of the total (developed + undeveloped) is disturbed when looking at the developed part separately. This is the part of reserves actively engaged in production, so one may expect some more dynamic effects. In F.2 we see the developed resources separately. These come to about 18 billion BOE at the end of 2015.
Superimposed on the reserves, we have over-plotted the cumulative drawdown on two of the resources: NG, and the total of all five products. We note that during the 6 year period in question, some 8.5 billion BOE were produced, as the total developed reserves went slightly up from 17 to 18 billion BOE. This shows an enormous effort to replenish, since production is on the same order of magnitude as the developed reserve. Also over-plotted is NG cumulative drawdown. While remaining reserve is some 7.5 billion BOE, cumulative depletion reaches 4 billion BOE. Again, same order of magnitude, requiring enormous effort to replenish NG reserves. Perhaps most significant here is the two step ramp-up of bitumen reserves. First, rise to 2 billion BOE 2012-2013; then, up from 2 to 4 billion BOE in 2014-2015.
The composition of the total developed resources is shown in F.3a and F.3b, for 2010 and 2015 respectively. Note that in 2010 NGL was likely small and reported with crude, but is reported separately and is quite significant in 2015 as shown in F.3b. What we clearly see in F.3, is that crude has shrunk appreciably, from 37% to 27% of the developed reserves mix, while bitumen increased from 3% to 23%. The largest shrinkage is in NG, which went from 56% to 41%, but still kept its place as the largest developed resource.
In summary here, the inability of Exxon Mobil to replenish 1.67 billion barrels of crude in developed reserves, over a period of just six years, must be appreciated. The very costly move into oil sands and bitumen would not have been made, were there any rational expectations of fresh crude resources. In other words, in 2010-2015 we have witnessed a new era, in which the most powerful oil company on the planet is unable to replenish its developed petroleum reserves.
2.1 Bitumen to the Rescue.
Whereas Exxon Mobil's effort to replenish petroleum resources by bitumen looks quite heroic, it is in the meantime expensive in CAPEX, risky, and brings but a low net revenue due to high OPEX. Risk comes from two directions which were never a serious consideration, until now. These are, (1) water availability, and (2) the environment.
For CAPEX, Consider for instance the two sites in Athabasca, Canada, owned and operated by Imperial Oil, Ltd., which is 70% owned by Exxon Mobil. These sites are, Kearl with Debottlenecking, and Kearl Expansion, both in the Fort McMurray area, and both use strip mining and the Clark hot water process for bitumen separation. Their capacities are, 235,000 and 110,000 BPD of bitumen respectively, and their CAPEX was $13.4 billion and $9 billion, respectively. These are quite expensive, relative to crude fields.
The hot water oil sands extraction process for strip mining requires about 4.5 barrels of fresh water per barrel of bitumen. At full capacity production, the Kearl operations would demand an amount of fresh water of 247,000 cubic meters of water daily, a huge amount. The Athabasca river, which until now was the source of the water used by the strip-mining bitumen producers, is already running quite low. In the meantime, 90% of the water ends up in toxic unlined settling ponds. The whole field of oil sands strip mining came recently under attack by environmental groups, due to the vast scale of land devastation it creates, and under severe pressure due to looming water shortages.
In contrast, the recent (2013) acquisition by Exxon Mobil of 226,000 acres at Clyden, from Conoco-Phillips (NYSE:COP), cost $720 million just for the acreage. Located in south Athabasca, the Clyden site is for in-situ bitumen extraction, by the steam assisted gravity drainage (SAGD) method. By my calculation, CAPEX for a new SAGD production capacity of 125,000 BPD of bitumen, would be $5.5 billion. Although less expensive than strip mining operations of the same capacity, the CAPEX here is by no means trivial.
In situ SAGD recovery requires about 1 barrel of fresh water per barrel of bitumen extracted. Although water usage is far lower than in the Clark method, water purification for re-use has created large quantities of toxic sludge.
In summary, even before touching the issue of (non) availability of pipelines for bringing the product to the Chinese market through the Canadian port of Vancouver, the whole field of oil sands bitumen production faces environmental challenges, very expensive remediation, and water shortages. As we will see later, the cost of production is high, and leaves a rather small net profit. It is risky, expensive, and unsustainable. In one word, an act of desperation.
2.2 Shale Gas Resources
Fast backward to December 2009, Exxon Mobil bought out XTO for $31 billion in shares, plus assumed debt of $10 billion. A total of $41 billion invested. If the Marcellus, Pennsylvania operations of XTO are any indication, this company is near the bottom of its class in performance, namely NG volumes (or MegaBTU) extracted per well. Since I have discussed this item in previous articles, there is no point repeating. The following observations are offered.
The recent monthly report available from the Marcellus (March 2016) shows XTO having 223 producing wells, active for 29 days on average, each producing on average 0.94 million standard cubic feet (SCF) per day of NG. In comparison, Cabot Oil & Gas Inc., a stellar performer, had for the same period 481 producing wells, working on average 30.5 days, and delivering on average 4.02 million SCF/day/well. We doubt this fourfold difference in performance between Cabot and XTO is due to an outstanding bad luck in the choice of well sites by the latter.
We note here that about $7.5 million is the cost per horizontally drilled, fracked, single well, in the Marcellus. Longer horizontal legs will cost more to drill, much more to hydro-frack. And there is still the need for some 5-6 million gallons of fresh water. These wells will not produce at a uniform volumetric rate; rather, a spike, followed by exponential decay, the tail of which may last several years. Observing 6,300 Marcellus wells taken at the end of 2015, (a mixture of few new and many old wells), I calculated an average of 2 million SCF/day/well. This is not brilliant, is quite expensive, and runs the risk of water shortage. Sounds familiar? And one more thing. Once started, there is no turning-off and waiting for a higher price in the market. The NG will re-adsorb into the shale, and the expensive fracking cost component (about $2.5 million per well or higher), will be lost. That is why the pipeline operators, or the buyers, (where well owners are unlucky enough to have their service) managed to drop the price of Marcellus NG to about $1/ MegaBTU. Lack of pipeline service prevents efficient pathways to markets in the US northeast, but should these networks be built, prices can be expected to drop further, due to competition.
In summary, for Exxon Mobil in particular there is no comfort in shale-NG in the Marcellus, or anywhere else. Furthermore, shale-based liquids, and condensate, are expensive to produce, have difficulty reaching the market for lack of proper safe transportation (high volatility), and do not get the high price deserved for a clean superior product.
2.3 Conclusions on Reserve Strategy
The 2010-2015 developed proved reserves actually increased slightly for Exxon Mobil. While NG and crude declined, the upward increment was due to an aggressive push into oil sands' bitumen, which went from 0.5 to 4 billion BOE in developed reserves, an 8-fold increase. Bitumen at the end of 2015 was full 23% of total developed reserves, a very significant proportion. The point we are making is that bitumen is an expensive, high CAPEX endeavor. Bitumen, especially strip-mining oil sands, is very risky due to water shortage, narrow gross margins, environmental pressures to cease and to remediate. The water issue, which is cardinal to bitumen production, makes bitumen essentially unsustainable. Therefore, due to the vulnerability to risk introduced by bitumen, the appearance of even-keel sailing, inferred by the stability of Exxon Mobil total reserves over 2010-2015, is totally misleading.
In a way of introduction, we take a look at Exxon Mobil's operating revenues from Upstream, Downstream, and Chemical, as reported on it 10 K submissions, for the years 2006-2015. This is shown in F.4. Note that progression is from right to left. Clearly until 2014, Upstream exploration and production, E&P is the leading earning mechanism. The other two activities do not distinguish Exxon Mobil from others, so we focus here on the Upstream component. Clearly, in 2015, a drastic loss of earnings from Upstream is evident: From $27.5 billion in 2014 to $7.1 billion in 2015: an 4-fold drop in one year. Of course we all know where this price drop originates, nothing Exxon Mobil could control.
Exxon Mobil's annual production volumes are summarized in F.5, over the years 2010-2015. We note that per unit energy, or BOE, production of NG and crude are very close. Recalling the developed reserves of the same two resources, in F.2 above, it is obvious that NG reserves are nearly twice those of crude.
The composition of the total volumes produced in 2010 and 2015 are shown in F.6.
The gross margins (price minus production cost) from sales of upstream products can be calculated from the prices, costs and respective daily production volumes reported. We have assumed 333 production days/year, which is congruent with the standard 8,000 hours/year of online operation. The calculated gross margins are depicted in F.7 for the same period. Note that bitumen cost of production is relatively high, so margins are far lower than for crude.
What F.7 reveals, is that gross margins for crude have started slipping already in 2011, and continue declining to-date. Also evident how small bitumen, NGL, and Synthetic contributions to the total are. The Price, and production cost, for each product is given in the 10 K reports. The calculated differences are called here specific gross margins per BOE of product, and shown in F.8. These data were used to calculate the values for F.7.
Note in particular the wide trajectory of crude margin in F.8, from 2009 to 2015. The calculated gross margin composition is further highlighted for 2010 and 2015 respectively, in F.9a and F.9b.
What is striking about the calculated compositions in F.9, is that they remain very similar for the respective components. This means that the sources of upstream income remain roughly in the same proportion. Note for instance bitumen, 2% in both 2010 and 2015, even though its volume proportion increased from 3% to 8%; this indicates bitumen is not an efficient revenue source. Otherwise, whereas crude has decreased somewhat, NG+NGL have increased. Also evident, as expected, that the total gross margin is halved between 2010 and 2015, from $66 billion to $32 billion.
The Upstream portion of total operational income, per F.4, has lost its dominant position after 2014, as its net fell from $27.5 billion to $7.1 billion. Within this (shrunk) Upstream domain, crude has largely retained its dominance as the largest earning mechanism. Wherein lies the problem of inability to effectively replenish reserves. If Exxon Mobil had no further additions to its crude developed reserves, these would last another 8.5 years or so, with crude production frozen at the 2015 level of 580 million barrels/year.
We are not saying that crude has disappeared, or is non-obtainable. Yet, world-wide demand for crude, regardless of GDP growth, will only increase, while global resources shrink, regardless of deep-sea "discoveries" already accounted for. Whence, uglier and hungrier sovereign toads must be kissed to get any, as years go by. For Exxon Mobil, crude has already started an earning decline trajectory in 2011, three years before the start of the Saudi Arabia-Iran war, on July 8, 2014. In a word, crude is already unsustainable.
4. The Financials
We will use here a physical measure to estimate a particular reserve lifespan. The particular proved reserve volume, divided by its annual rate of production, should give an estimate of the years remaining-if both reserve and production rate remained unchanged. Of course this precludes the effect of diligently replenishing of the reserve, and includes the unhappy case of inability to replenish at reasonable cost. In other words, this is a conservative estimate of the actual lifespan. This calculation is repeated at the end of each year, so the estimated lifespans may vary for the same resource according to the levels recorded at year's end. Now in this approximation, we use only the developed reserves, assuming 100% are available for production, ignoring the unproved reserve portion. This calculation is repeated for crude, NG and bitumen reserves, and the results shown in F.10. With the exception of bitumen, crude and NG lifespans are confined to 13 years or under. Bitumen developed reserve sizes are very large compared to the amounts withdrawn, so lifespan goes to over 40 years.
To continue this simple exercise, we would calculate a net present value of the particular resource at the end of each year, by assuming that the year-end gross margin multiplies the annual production reported, to provide a payment. The constant withdrawals and payments over the calculated life-span are used in a net present value calculation. We assume an annual discount rate of 10%, uniformly for all 3 resources, so we denote the calculation as PV10. Of course, we do not have the CAPEX directly associated with these assets, so our calculated PV10s would tend to the high side. The annual PV10s calculated this way for the 3 developed resources of crude, NG and bitumen, and their total, are drawn in F.11.
Over-plotted on F.11 is the total assets value of XOM, per its 10 K submissions. It is emphasized that the important observation is the order of magnitude closeness, and the trend shown by the black line denoting total of the 3 resources. This PV10 calculation has a downward trend since 2011, way before July 2014. It differs from the accounting version of total assets, which is moderately increasing from 2010 to 2015. We hope that the net present value of these assets really does not go much lower than their book value, or else impairments are looming ahead, an undesirable specter.
A look at financial performance as reported, is shown in F.12. Net Income, long term debt, total debt, and dividend payments are plotted. Already in 2012 we see a sharp decline in net income, coupled with a sharp rise in total debt, and long term debt, as one would expect from the even-keel sailing. It may be interesting to point out that as the net income declined, dividend payments were also increased, incrementally. In 2015, total dividend payments were $12.08 billion, while the net income was $16.15 billion. Shareholders were getting a local anesthetic, while Net dropped some $30 billion, losing 2/3 of its 2012 value.
We will not go into the Exxon Mobil stock value, which is likely driven mostly by a variety of high frequency trading algorithms, vying through favorite electronic exchanges to trip-up and better-buy, or out-sell each other, at a frequency of over 10,000 flops per second, completely without human intervention. For these algorithms, only volatility matters.
We will also leave alone the speculation regarding the future of oil prices, which is clearly in the hands of Saudi Arabia. My speculation: If, and when, some form of non-aggression agreement between Saudi Arabia and Iran is reached, only then the oil spigot will close, and oil prices will shoot up again. Now what happens to the Saudi fields when they are producing at a rate of 10.5 million BPD for 2 years? Accelerated decline. So world oil reserves will be tighter, while demand continues to rise. Whereas artificially-high oil prices in 2008 caused havoc with the US work force, the low oil prices of today fail to bring comfort. If you are poor and working, $45 more in your pocket saved on gasoline will not save the month, but $45 less, as in 2008, can be devastating.
Somewhere Exxon Mobil hopes that the day of very high oil prices returns, and is nigh. But in a new world of rapidly shrinking resources, the former strategy of upstream reliance will leave it very vulnerable. Most of all, Exxon Mobil reliance on bitumen and shale-gas is unsustainable.
5. What Should Mr. Darren W. Wood Do?
5.1 become totally sustainable in 5 years, abandoning crude entirely.
Currently, there are 13 billion gallons of corn-based ethanol on the market. One gallon of ethanol comes via fermentation from 2.8 bushels of corn. If you grow on average 149 bushels/acre of corn, the total area in ethanol-corn is 31 million acres. Regarding net carbon, corn ethanol is a total failure. Fully 37% of total US corn acreage are for ethanol. Over $3.6 billion are paid annually as government subsidies. Large amounts of nitrogen are used to get high corn yields, resulting in runoff which polluted the gulf of Mexico. The 99% starch corn is not a food, but a chemical feedstock.
Now growing switchgrass, and miscanthus, and energy cane (a US invention) can yield about 10 dry Tonnes/acre/year. Compared to corn, these are easier to grow, need far less fertilizer, herbicides, pesticides, and less water. This material can be gasified to make syngas, and then made into high octane gasoline via Exxon Mobil's MTG method (methanol to gasoline, 1972 patent), quite economically. Not a dream, I have done it with my own hands, and built a successful pilot in New Jersey, 2007-2012. The yield? 2.5 Tonnes or 21.3 barrels of gasoline/acre/year.
Now if you cultivate 31 million acres, you can grow high octane gasoline, some 660 million barrels/year. This grows back, every year. This volume of finished product is equal to Exxon Mobil's 2015 production of crude plus NGL combined. The projected costs would be lower than those associated with bitumen production.
The foregoing is provided for reference only, and can be carried out with a proper strategy regarding growing, harvesting, densification, and intermediate small scale processing. One may make synthetic gasoline, or a higher volume of methanol, some 1.9 billion barrels/year in this method, and distribute it directly into cars. Today's gasoline auto engines have a software switch allowing use of 100% methanol, which is an oxygenate, and actually burns better than gasoline. Of course there is a difference in specific energy, methanol has about 70% of gasoline's energy per unit volume. Still, a respectable range from a tankful, low emissions, and, zero net carbon.
What is better than crude? To have the clean automotive product instead. It would take some four years to fully cultivate crops and develop all subsystems and logistics. It takes a very strong corporation to make such a sharp turn, and carry out this strategy. Yet the rewards are very rich: a sustainable resource, in continuous demand, at a price determined by market supply and consumption, not a remote despotic regime. As bonus, Exxon Mobil can become an environmental hero, not the villain it is portrayed today.
In conclusion, Exxon Mobil in 2016 seems at a standstill. Its fossil reserves are difficult to replenish, or, for bitumen, introduce high risk. The prices for its key products are determined by remote entities, not by market logic, and are currently down for 2 years. Its recent increased reliance on downstream, a highly competitive market, has its own risks, upon price recovery.
Only a bold move can turn this tide. Mr. Wood has just the right name for it.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.