1) BLU PUTNAM, CHIEF ECONOMIST, CME GROUP JANUARY 2016 Oil Market Dynamics and 2016 Outlook All examples in this report are hypothetical interpretations of situations and are used for explanation purposes only. The views in this report reflect solely those of the author and not necessarily those of CME Group or its affiliated institutions. This report and the information herein should not be considered investment advice or the results of actual market experience. In the second half of 2014, the oil market experienced a even as oil prices were falling – and see what lessons can powerful downward price adjustment which was sustained be learned about the behavior of suppliers in a low-price throughout 2015. There are several long-term supply environment. That is, there are short-term and long-term and demand forces in play, as well as some shorter-term feedback loops from prices-to-production decisions that are response factors, that make for a very difficult mix to analyze quite complex, and that worked to reinforce lower oil prices, going forward. On the supply side, there are the technology- at least in 2015. Finally, we cover an important change in driven improvements in extraction techniques that ignited market structure – namely, the lifting of the U.S. ban on a production boom in the United States back in 2006. On crude oil exports – to appreciate some of the dynamic shifts the demand side, there is the huge shift in the global growth involving relative price spreads inside the workings of the environment from an emerging market boom period in the broader global oil market. As suggested above, both our early 2000s to a sluggish growth period after the 2008-2009 long-term and short-term analyses place a high probability Great Recession. Also, technology has been steadily making on the base case that the low crude oil price environment transportation considerably more fuel efficient. Shorter- has many years to run. There are, however, a number of very term factors include the time-lagged feedback loops and low probability scenarios that could cause a return to higher behavioral responses to producers adjusting to a lower oil prices, which deserve at least some passing consideration. price environment, as well as policy responses such as the lifting of the U.S. ban on crude oil exports. Figure 1. Our forward-looking analysis of the long-term trends in the crude oil market, including sluggish global growth, continued advances in transportation fuel efficiency, and extraction technology improvements, suggests that the era of relatively low prices could last for many years. In hindsight, what seems remarkable is that oil prices stayed as high as they did for as long as they did, before breaking down in the second half of 2014. This research report starts with a brief review of the three key long-term trends we perceive are the main drivers of crude oil prices, and we assess the probabilities of any shifts in these trends. Second, we take a look back at the perceived catalyst that “broke the camel’s back” – namely, the Organization of Petroleum Exporting Countries’ (OPEC) decision in November 2014 to keep producing at high levels 1 

2) JANUARY 2016 I. CRITICAL LONG-TERM TRENDS ARGUE FOR A SUSTAINED LOW PRICE ENVIRONMENT As noted in the summary, the oil market faces both demand and supply trends arguing for the continuation of low prices. The two persistent demand trends are (1) slow global growth and (2) continued advances in transportation fuel efficiency. The long-term supply trend is grounded in the pace of technological improvements related to continued reductions in extraction costs. None of these factors seem likely to change course over the next several years and some may be accelerating in intensity. dependent country. With its major trading partners no longer posting solid growth, exports have stagnated. Importantly, none of these four key factors pointing to a slower growth rate are reversible by short-term government policy adjustments. Indeed, China is experiencing a natural process of reduced long-term growth potential as a result of the success of its modernization programs, while dealing with slow-moving demographic patterns and lackluster global growth. Policies such as devaluing the currency are not likely to help much on the growth front in the short-run, even if a depreciating exchange rate against its trading A. Slow global growth. China is decelerating. Brazil is in partners is the most likely scenario for the currency going recession, as is Russia. Most emerging market countries forward. We are not forecasting a hard-landing for China, but are struggling to grow. In mature, industrial countries such just a very bumpy road to a 3% real GDP trend growth rate in as the United States, Europe, and Japan, about the best the 2020s, which is neither supportive of higher commodity that can be expected is 2% real GDP growth, and even prices in general, or oil prices specifically. that low bar may be tough to achieve in 2016. In essence, the era of strong commodity demand in the early 2000s that was supported by 10% real GDP growth in China, and strong growth in many emerging market countries is long gone, with little prospect of returning. And, in the post-2008 environment, mature, industrial economies are struggling to produce anything better than lackluster growth. The implications of this outlook are for very sluggish growth in energy demand, especially for crude oil. And, it is not clear what could change the growth prospects. There is a secondary effect related to China’s impact on many commodities beyond oil. China’s shadow banking system has depended heavily on using commodities as collateral for lending. During the China super-growth period, the use of commodities as collateral meant that commodity demand was accelerated beyond that associated with rapid economic growth.1 The reverse is true, too. As China has decelerated and commodity prices have fallen, some collateral has been released into the market and the demand for new commodity-collateral for lending has fallen, China’s growth deceleration is based on four key factors. making the China impact on commodity prices even more First, the country grew rapidly as it invested in infrastructure pronounced than the economic deceleration might suggest. at an impressive pace in the years from 1980-2010, but with modernization has come the reality of diminishing returns from new spending on nation-building projects. Second, the aging demographic pattern is making a transition to a more domestic-demand driven growth model extremely difficult. The over-65 age group will make up more than 20% of the economy in the 2020s, and retirees spend considerably less per person than working age individuals. Rolling back of the one-child policy in 2015 may help ease the demographic challenges by 2050, but not over the next decade or two. It takes thirty years to make a thirty year old and materially impact the growth of the labor force. Third, China has benefited from a large rural-to-urban migration pattern that has supported growth for several decades. As the percentage of the rural population declines in the 2020s, though, this source of growth for the economy will diminish as well. Finally, we note that China is still an export1 2 The U.S., Europe, and Japan have all tried every manner of expansionary monetary policy to pump up growth since the 2008-2009 recession but to no avail. The lack of success of monetary policy to create superior growth was because it could not address the fundamental reasons for slow growth. 2 The first and largest challenge to growth potential in these mature, industrial countries is the demographic pattern. Populations are not growing, and are aging while labor force growth is next to zero. On the demand side, as already noted for the case of China, per capita consumption spending declines for the retired demographic, and this is the faster growing segment of the population. With respect to potential GDP growth, if there is little to no labor force growth, then it takes above-average increases in labor productivity to create superior growth rates. While this is possible with technological gains and out-sized capital Shaun K. Roache and Marina Rousset, "China: Credit, Collateral, and Commodity Prices", Hong Kong Monetary Authority, HKIMR Working Paper No.27/2015, Fall 2015. Bluford H. Putnam, "Essential concepts necessary to consider when evaluating the efficacy of quantitative easing." Review of Financial Economics 22.1 (2013): 1-7. 2 

3) JANUARY 2016 investments, it is highly unlikely in mature economies. We give this possibility a meaningful 25% probability, and it Indeed, without major tax and labor market structural would most likely point to a temporarily lower level of global reforms, which are only remote possibilities, it seems unlikely crude oil prices. Second, supply disruptions are possible that sustained, above-average labor productivity gains are depending on how tensions in the Middle East, particularly possible. Low short-term interest rates and central bank relations among Saudi Arabia, Iran, and Iraq develop. At the asset purchases (i.e., quantitative easing) can raise asset present time, we place the probability of military action that prices above what they might have otherwise been, but there would impair global supply as very small, less than 10%, is little to no evidence that they can raise labor productivity. but worthy of close monitoring as this low probability event Recession or slow growth in most of the emerging market countries are not so easily grouped into one cause. Commodity producing countries are naturally suffering would come with huge price action potential to the upside. II. APPRECIATING THE CATALYST FOR THE PRICE BREAK in a slow-demand world. But political risks are high and When long-term supply and demand forces are in play, it can rising in a number of countries, from Brazil to Turkey. take markets many years to realize their full impact. Often, And, among the Asian countries close to China, it is the there is a catalyst for a price break which can be incorrectly deceleration of their big neighbor that dominates growth interpreted because of its association in timing with a critical prospects. Regardless of the cause, it seems hard to project event that may have been building for years. We argue that resurgence in emerging market countries’ growth without this was the case in the fall of 2014. either stronger growth in China or in the mature, industrial countries -- and neither are in the cards for 2016. Figure 2. B. Fuel Efficiency. Also on the demand side, and not always given its proper due, is the continued progress in transportation fuel efficiency. Crude oil is 70% to 75% a transportation fuel in terms of the uses of refined petroleum product. The relentless march toward greater fuel efficiency is both impressive and a continued drag on crude oil demand growth. The elasticity of demand for crude oil with respect to real GDP growth is a on a long-run declining path. Indeed, there appears to be considerable gains still possible in the fuel efficiencies of internal combustion engine vehicles, including the ability to use lighter materials, such as aluminum in pickup truck frames. Natural gas is starting to make inroads into transportation in bus fleets, in long-haul truck transport, and in railroad engines. Electric vehicles are still a minute portion of the transportation system, but longer-term advances in battery technology could change that, especially if batteries can be made both lighter as well as more efficient. And, the promise of cleanburning hydrogen, with H2O coming out the exhaust pipe, remains a long-term dream attracting considerable research and development funding. C. Risk Factors. While the base case of slow global growth has a very high probability associated with it, there are risks. First, global growth could be even slower, almost stagnant, if China’s deceleration turns into a hard landing. Crude oil prices had been bouncing around $100/barrel all through the first half of 2014. Prices started drifting lower in July 2014, and broke below $90 in October 2014. In midNovember, prices broke below $80. On 27 November 2014, OPEC met in Vienna and decided to maintain production rather than cut back and try to support the market. “Recording its concern over the rapid decline in oil prices in recent months, the Conference concurred that stable oil prices – at a level which did not affect global economic growth but which, at the same time, allowed producers to receive a decent income and to invest to meet future demand – were vital for world economic wellbeing. 3 

4) JANUARY 2016 Accordingly, in the interest of restoring market equilibrium, One has to appreciate the difference between cash flow and the Conference decided to maintain the production level of accounting reports. Oil producers, just like mom-and-pop 30.0 mb/d, as was agreed in December 2011.” convenience stores, know that cash is king. It is very easy 3 The OPEC statement, emphasizing the need for oil producing countries to keep on producing to maintain their incomes, fueled concerned that oil prices would fall further, and fall they did. to include non-cash items, such as depreciation, as well as certain investment or capital costs, into the calculation of how much money is required to produce the next barrel of oil. What really matters to the oil producer, though, are the actual cash costs of the next barrel of oil, and these cash During 2015, much was made by many analysts about the costs can be substantially lower than the costs as measured Saudi Arabian strategy to increase production to squeeze by accounting principles and purported to represent the out high marginal-cost producers. We would argue that dollar-cost per barrel.5 What this means is that on a going this is much more rationale than strategy. That is, the key forward cash basis, production that looked unprofitable on problem for Saudi Arabia, not to mention virtually every an accounting basis was still net-cash-flow positive or only other OPEC member, is that their government spending small cash losers. So the producer kept on pumping – oil programs were built on assumptions of crude oil prices and cash. staying above $80 per barrel (or perhaps a little higher) for as far as the eye could see. $40 per barrel oil puts tremendous domestic political risk into the equation for the governing authorities, as it makes it nearly impossible to continue with planned spending and subsidy programs without material adjustments. Indeed, Saudi Arabia has gone to the debt markets for new money as well as cut Debt matters, too, because many oil producers have a lot of it. If they were to shut their production down, there would not only be no flow of oil, but also no cash flow; and the cash is needed to pay their debts. Pumping oil at a loss makes sense if one can stay in the game for the long-run and avoid bankruptcy. back government spending and subsidies. By increasing And then there are the advances in technology. Producers production, they were able in some small way to keep the in the United States using hydraulic fracturing methods cash flowing. As we analyze the behavioral response and and horizontal drilling have been increasingly improving feedback loops to lower prices in the crude oil market, we their ability to finish wells faster and get to the new wells will again come back to the theme that long-term spending sooner, at lower costs. “Finishing faster” simply means and liability commitments work to keep production strong, sharply increased production in the first months of a well’s even with low prices, at least for an extended period of life, allowing the producer to close the well sooner and move time, which creates complex lags in the price-to-production to the next location. Rigs are now available that can “walk” response cycles. (albeit very slowly) to the next drill site. Enhanced fracturing techniques can improve extraction results. It all added up III. BEHAVIORAL FEEDBACKS AND POLICY RESPONSES IMPACT CRUDE OIL SPREADS A. US production dynamics. When crude oil prices collapsed by half in the fourth quarter of 2014 many analysts expected a relatively quick supply response based on models taught in every Economics 101 class. Unfortunately, the basic economics version of supply and demand knows nothing about debt, time, and cash flow, among many other things. Many wells in the United States were shut down in 2015 as predicted, but a focus on getting more oil from the most efficient wells kept production higher than most analysts expected.4 There were a couple of reasons for this common miscalculation. in 2015, allowing U.S. producers to cut rig count and still maintain strong production. What may change in 2016 is the lagged impact of sharply reduced capital investment in 2015. That is, while producers were using technological improvements to get more oil from fewer rigs, they were not investing in new capabilities. Virtually every capital investment project that could reasonably be delayed in 2015 was, indeed, delayed or postponed indefinitely, while oil producers assessed their economic future in a lower price environment. By the second half of 2015, most producers had come to realize that that they were in for a very long period of lower prices, and the process of downsizing and consolidation began in earnest. “OPEC 166th Meeting concludes”, Press Release No 7/2014, Vienna, Austria; 27 Nov 2014; http://www.opec.org/opec_web/en/press_room/2938.htm 3 James Hamilton, Professor of Economics, University of California at San Diego, presentation on “Fracking, China, and the Geopolitics of Oil", at the Research Council, J.P. Morgan Center for Commodities (JPMCC), University of Colorado @ Denver, December 4, 2015. 4 Bluford H. Putnam, “Oil Price Lessons from 1983”, Euromoney, December 15, 2014., and Bluford H Putnam and Desiree Schwartz, “Visualizing Energy Market Dynamics” December 4, 2014, CME Group, also reprinted in the Hedge Fund Journal, December 2014. , 5 4 

5) JANUARY 2016 We draw two conclusions from the mix of improved 1. What has really changed? extraction technology and lack of new capital investment. Under the old law, U.S. refined product was allowed for First, when oil prices rise, at some threshold price probably export. Crude oil exports required licenses. Effectively, north of $50 per barrel yet well below $80, new production U.S. crude oil could be exported to Canada and Mexico will again be profitable and will come online faster than in the with permits, which were virtually automatically granted, past. Second, the lack of capital investment in 2015, and the as were re-exports of foreign-sourced oil, and some crude likely lack of investment in 2016, will start to hit production, oil exports from California and Alaska. Moreover, the such that U.S. production in 2016 and beyond may show definition of refined product had been weakened in the material declines from 2015 and 2014 levels so long as oil last several years to include some lightly altered crude prices are below $50 per barrel. This capital investment products (i.e., lighter condensate products). With the impact may be much larger for higher cost production lifting of the crude oil ban, U.S. producers now can export areas, such as the North Sea, which has seen declining freely; however, do not expect much of a rise in exports of oil production for over a decade and has higher on-going crude oil any time soon. Indeed, the sum of crude oil plus maintenance and marginal costs. refined product exports is likely to remain more or less on From a global supply picture, the reduced production in 2016 from the United States, the North Sea, and probably its current trend for 2016-2017. 2. Will lifting the crude oil export ban result in greater Alberta, Canada, is not likely to impact prices since it may be U.S. production? more or less offset by rising production entering the world No. The low price environment for crude oil globally that market from Iran. It is unclear how the balance will tilt, but commenced in Q4/2014 is still with us, and the longer- the overall impact on prices may be more to cause short- term expectation for price is the key driver of future term volatility within a wide price range rather than to push production. As noted earlier, China is still decelerating. prices back on to a sustained rising trend. Growth in emerging markets is slow. Europe, U.S., Japan B. U.S. lifts oil export ban. The U.S. crude oil export ban was lifted in December 2015 as part of the legislation to fund the Government through September 2016. The export ban was imposed back in 1975 under the administration of President Gerald Ford in the midst of public anxiety over the rising power of OPEC, reduced US influence over global economic conditions, and fears of slow growth and high inflation – then known as stagflation. In fact, through Presidential actions over the years and other rule changes, the ban was quite leaky, so to speak. As a result, the short-term impact on oil prices of lifting the export ban is likely to be relatively small in terms of prices and not an important driver for production. may grow 1% to 2% in real GDP terms. No major demand surges here. And as discussed in the technology trends section, oil is largely a transportation fuel. Transportation is becoming steadily more energy efficient. In short, as we have argued, the demand situation does not support a return to a higher price environment whether the U.S. exports oil or not. Nevertheless, the lifting of the U.S. crude oil export ban will mean some small benefits to U.S. producers based on the tighter Bakken-WTI spreads, because Bakken and other domestic sweet crudes will now have new export markets that will bring higher revenue overall. 3. What is the likely impact on Brent-WTI and other crude Nevertheless, anytime frictions and barriers to free trade oil price spreads? are removed, the market price discovery process is made Our view is that any policy change that removes market more robust and capital allocation more efficient. Hence, frictions and makes the connection among different the lifting of the export ban is a positive factor for the role sources of oil around the world more efficient will of U.S. oil (West Texas Intermediate, aka WTI) as a global assist the robustness of the global oil price discovery benchmark. Here we provide our perspective on some of the process. Thus, the lifting of the U.S. crude oil export ban key questions being asked. could make an incremental difference in narrowing the spread between North Sea Brent and U.S. West Texas Intermediate (WTI). 5 

6) JANUARY 2016 Figure 3. to nothing over the coming years; however, it can still be quite volatile around the average given the potential for weather and maintenance supply-disruptions in the North Sea, and lower production of Brent point to much higher basis risk. 4. What are the implications for refined product? Another implication of lifting the ban is that U.S. oil exports will go where the combination of lower transport costs and refinery demand coincide. This probably means some increase over time in U.S. oil exports to Asia. Remember though, crude oil is not so much exported to a country as to a refiner that happens to be in another country. So, the direction of crude oil exports from the U.S. will depend on developments in refining capacity in the U.S. and around the world, as well as on Indeed, in early December, as news of the possibility that the crude oil export ban might be lifted, there were some price actions between different grades of crude. transport costs between competing sources of crude oil. U.S. refineries are extremely cost efficient. They are more than capable of competing effectively around The WTI-Brent spread narrowed slightly in both spot the world – again putting increased emphasis on and longer-dated futures. Also, the Bakken-WTI spread transportation and storage costs. Also, in virtually tightened. In addition, in the U.S. Gulf Coast, WTI and LLS all countries around the world, it is difficult to get the (Louisiana Light Sweet crude) went up in price relative permits to build new refineries – not impossible, just to the sour crude grades (e.g., Mars). The export ban hard. So, to build a new refinery, one needs billions of artificially depressed sweet crude in the U.S. Gulf Coast dollars, a steady source of foreseeable demand, and market relative to sour crude. So, there was a perceptible reasonable transport costs to get the crude oil from the change in the sweet-sour spread in the U.S. Gulf Coast source to the new refinery. Put another way, changes market in the weeks prior to the lifting of the crude oil export ban. in the global environment for refineries will drive some Prior to 2006 and the emerging U.S. oil production meaning U.S. refined product exports will probably hold boom, WTI and Brent met, in a competitive sense, up quite well even as there may be incremental increases at the refineries in the northeastern United States. in crude oil exports. From 1993-2006, the Brent-WTI spot price spread was typically extremely narrow and not very volatile. With the production boom, higher U.S. oil production eventually overwhelmed capacity to deliver oil where prices were higher and refined product exports had not yet taken off. During 2011-2012, Brent was consistently priced $20 above WTI, with a peak of $29.70 in September 2011, while markets were temporarily separated. In 2013 and shift in exports over the coming years, but not quickly, Long-term, as the competitive landscape for refineries adjusts, there may be some incremental narrowing of refined product price spreads relative to crude oil since the crude oil market will be just a little more efficient. This will take time and may turn out to be of relatively small impact. 5. What is the state of the U.S. infrastructure for 2014, the oil delivery and storage infrastructure in the exporting crude oil? United States largely caught up with greater production, The lifting of the export ban will have the biggest impact and the spread has since narrowed materially. Since in the U.S. Gulf Coast, and to a lesser extent on the West lifting the crude oil ban incrementally improves the Coast/Alaska. The infrastructure for WTI exports in the competition among all sources of oil, it suggests that U.S. Gulf Coast is already completed, and the United the spread between Brent and WTI will average next States is actively exporting some crude oil and lighter 6 

7) JANUARY 2016 condensate products, not to mention all the NGLs such as propane (which use the same export terminals). At this time, there is adequate capacity to handle any increases in export flow. Indeed, in January 2016, the first boat loaded with U.S. crude was leaving Texas for Europe. IV. SUMMARY Our analysis points to a base case for long period of low crude oil prices. The China-driven boom in emerging market demand is over. Slow growth due to aging demographic challenges will keep real GDP growth very slow in the mature, industrial countries. Technological advances in fuel efficiency in transportation mean the elasticity of demand for crude oil with respect to economic growth is diminishing. And, on the supply-side, further technological improvements in oil extraction are reducing costs, allowing more production from fewer oil rigs. This all adds up to a prolonged period of low oil prices as our base case. The price risks to the downside for oil prices come mostly from the possibility of a hard-landing in the Chinese economy leading to a global recession – not likely but worth considering. The price risks to the upside come from conflict in the Middle East, possibly involving Saudi Arabia and Iran, leading to major supply disruptions – again, this is a small probability event with a huge impact, so monitoring is required. Copyright © 2016 CME Group. All rights reserved. 7