EFTA01097655.pdf

Deutsche Bank Markets Research VA United States HY Corporate Credit Date 13 January 2015 Energy Kathryn O'Connor Research Anent HY E&P Sector Update & 2015 Outlook Jaloci Research Associate Quarterly Updates for E&Ps under coverage including: In•Depth Relative Value Forecasts Sector Benchmarking Scenario Analysis E&P Universe Credit and Operational Metric Screen across 30 credits including: Results and Ranking Snapshot Scatter Plots by Metric Liquidity Analysis Scenario Analysis Historical Financial Summaries Selected Recommendations: BUY EXCO Resources (XCO) BUY Cimarex (XEC) BUY Halcon Resources (HK) BUY Hilcorp Energy I (HILCRP) BUY Newfield (NFX) BUY Range Resources (RRC) SELL Denbury Resources (DNR) Recommendation Changes: HOLD Chesapeake Energy (CHK) from BUY SELL Forest Oil/Sabine (FSTO) from HOLD SELL QEP Resources (QEP) from HOLD SELL SandRidge (SD) from HOLD SELL Samson Resources (SAIVST) from BUY E&P Screening and Analysis for the following E&Ps: American Energy Permian (AEPB) Newfield (NFX) Antero Resources (AR) Oasis Petroleum (OAS) Bonanza Creek (BCEI Parsley Energy (PE) Chesapeake (CHK) Penn Virginia (PVA) Cimarex (XEC) QEP Resources (QEP) Concho Resources (CXO) Range Resources (RRC) Danbury Resources (DNR) Rosetta Resources (ROSE) Diamond Back Energy (FANG) Samson Resources (SAIVST) EP Energy (EPE) Sanchez Energy (SN) Exco Resources (XCO) SandRidge (SD) Forest Oil/Sabine (FSTO) SM Energy (SM) Halcon Resources (HK) Triangle Petroleum (TLW) Hilcorp Energy (HILCRP) Tullow Oil (TLW) Magnum Hunter Resources (MHR) WPX Energy (WPX) Midstates Petroleum (MPO) Deutsche Bank Securities Inc. DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1. MCI IP) 148/04/2014. EFTA01097655 13 January 2015 HY Corporate Credit Energy HY E&P Sector Overview: Sifting through the carnage 3 E&P Credit Screens and Analysis E&P Credit & Operational Metric Screen 18 Overall E&P Screen: Results and Ranking Snapshot 20 Leverage Screen 23 Current PV-10 to 2016 Debt Screen 26 Liquidity Screen 28 Hedging Screen 32 Adjusted Cash Margin Screen 35 Commodity Mix Screen 39 Conclusion: E&P subsector weighting: Maintain Underweight 41 DB Commodities Price Deck & Outlook 42 E&P Industry Relative Value Sheets 46 Quarterly Updates for E&Ps under coverage Chesapeake (CHKI 54 Cimarex (XEC) 60 Denbury Resources (DNRI 66 Exco Resources IXCO) 72 Forest Oil/Sabine (FSTO) 77 HaIcor) Resources (HK) 82 Hilcorp Energy (HILCRP) 87 Newfield (NFX) 92 QEP Resources (QEP) 97 Range Resources (RRC) 102 Samson Resources (SAIVSTI 107 SandRidge (SD) 112 Financial Summaries and Sensivities for Non-covered E&Ps American Energy Permian (AEPBI 118 Antero Resources (AR) 120 Bonanza Creek (BCEI 122 Concha Resources (CXO) 124 Diamond Back Energy (FANG) 126 EP Energy (EPEI 128 Lightstream Resources (LTSCN) 130 Magnum Hunter Resources (MHRI 132 Midstates Petroleum (MPO) 134 Oasis Petroleum (OAS) 136 Parsley Energy (PE) 138 Penn Virginia (PVA) 140 Rosetta Resources (ROSE) 142 Sanchez Energy (SN) 144 SM Energy (SM) 148 Triangle Petroleum (TLW) 150 TuIlow Oil (TLW) 152 WPX Energy (WPX) 154 Page 2 Deutsche Bank Securities Inc. EFTA01097656 13 January 2015 HY Corporate Credit Energy HY E&P Sector: Sifting through the carnage Oil price collapse - how did we get here? Looking back, oil fundamentals at the recent peak in the June/July 2014 timeframe were seemingly decent. Despite some volatility, discussions with E&P management teams were bullish in the face of increasing service costs, and capex budgets looked to be rising as higher than anticipated oil prices drove greater than expected cash flow. On the geopolitical side, we had rising tensions in the Middle East posing a headwind for OPEC supply (driver of a majority of future OPEC supply growth). Further, M&A chatter was relatively strong after the Whiting (Will/Kodiak (KOG) acquisition announcement. In summary, there were few reasons the market had to be concerned about the sector. Fast forward six months and the world has changed, leaving investors to ask the question, "how did we get here?" Looking through the wreckage now, it does seem that there were missed signs that when taken to together pretty clearly mark how we reached our current oil destination (WTI trading at less than $50). From 2002 to 2012, global demand growth averaged over 1.1 million b/d while non-OPEC supply growth averaged only 560 Kb/d; this excess demand supported a generally increasing oil price. That trend reversed itself starting in 2013 as average global demand fell below 1.0mmb/d at the same time non-OPEC supply growth accelerated to an average of t6 million b/d. Specifically and more dramatically in 2014, the difference between incremental annual world demand growth and non-OPEC supply growth gapped out to an oversupply of 1.2 million b/d - a differential not witnessed in decades (Figure 1). This was the backdrop as OPEC met on November 27th and decided to maintain its ceiling production level of 30.0 million b/d. Figure 1: World Oil Demand and Non-Opex Supply Growth 2000 •World demand growth (kbd) 1800 Total non-OPEC supply growth (kbd) 1600 1400 - 1 I.[ II 1200 1000 - 800 600 - 400 • 200 - 0 2014E 2015E 2019E 2020E &wet BA a DB arntnAta Looking at the individual pieces of this story, the current outcome now seems like a reasonable one. Anecdotally, North American E&Ps had seen a relatively steady pattern of hitting production guidance if not beating - and raising - through the Q2 14 earnings season and eventually the Q3 reporting season. This came despite some tough weather to start off the year. Looking at the Deutsche Bank Securities Inc. Page 3 EFTA01097657 13 January 2015 HY Corporate Credit Energy demand side of the equation, there was a key difference in expected demand momentum as we moved through 2014 versus 2013 (Figure 2). While in 2013 there was a relatively slow but largely steady improvement in expected demand through the year, 2014 started strong but saw a marked deterioration in expected demand that started in August but declined precipitously as we moved through year-end. These two factors left a bit of a perfect storm for oil in and around the time of the OPEC meeting. While analysts were somewhat split on the expected specific outcome, a majority expected an OPEC production cut of some sort to help stem oil oversupply (9 out of 13 street analysts polled by Reuters). All of this led to a dramatic move in oil when OPEC decided to maintain the current 30.0 million b/d ceiling production without even a minor cut, which was seemingly the market's last chance to stabilize oil prices. Worse, all of this played out against an actual OPEC average production level 30.2 million b/d YTD through November 2014; the whisper production for December is 30.7 million b/d — and that includes lower Libyan production. Figure 2: Global Demand Growth Expectations Over Time mb/d 2013 — — 2014 2016 1.5 1.4 1.3 • • 1.2 • 1.1 • 1.0 • 0.9 0.8 • 0.7 S. • 0.6 ro 7 co 7 7 co e) m co • 7 • •7 IT 7IP I—V, a RI 0' # C I4 LL4 2 4 . asc Sane! 'EA 08Cooly Absturcet Oil looking forward - Are we there yet? The simple answer is no. Despite the significant amount of pain inflicted already, unfortunately, we believe there is likely more to come in 2015. Excluding the 2008/2009 (which was even more extreme), the average sell-off in an oil bear market reached its nadir with an average decline of 500/0 from the peak over a 15-month period (Figure 3). Currently, we have observed —60% decline in oil prices from the peak; however, we are only 7 months into this episode. Interestingly, of the four episodes shown, the one where oil reached its low point in the shortest amount of time was in 1986 - the historical data point that sets up the most like current scenario; in that case, OPEC too played a significant role in the 1986 oil bear market. However, at that time, it was other OPEC producers that were squeezing out marginal Saudi Arabian barrels. This time the US is playing the role of Saudi/OPEC spoiler by growing its unconventional production base and squeezing out OPEC barrels. In any case, despite a significant decline in oil prices so far, we don't see any expected changes to the major drivers of oil price in the near term. Page 4 Deutsche Bank Securities Inc. EFTA01097658 13 January 2015 HY Corporate Credit Energy IFigure 3: Prior Bear Markets in Industrial Commodities Commodity Stan End Duration, mo Drawdown, % Oil 11/2211985 07/25/1986 8 -56 Oil 10/18/1992 12117)1993 14 -37 Oil 01/10/1997 12/11/1998 23 -58 Oil 11/24/2000 01/1812002 14 -48 Nat Gas 10/24/1997 08/28/1998 10 -52 Nat Gas 04)06/2001 01/25/2002 10 -61 Nat Gas 06110/2011 04/13/2012 10 -60 Aluminum 08126/1988 02/02/1990 18 -45 Aluminum 08118/1995 03/12/1999 43 -38 Copper 08121/1992 10/22/1993 14 -37 Copper 08121/1995 02/22/1999 43 -54 Iron Ore 08126/2011 09/07)2012 13 -50 Avg for oil 15 -50 Avg tor all 18 -50 Current episode Oil 06/20/2014 01/10/2015 7 -54 Aluminum 04/29/2011 01/10/2015 45 -36 Copper 07/29/2011 01/10/2015 42 -35 Iron Ore 12/062013 01/10/2015 13 -49 Sane oe Oral' $ea411Y While recent oil downturns have had a relatively quick "V" shaped recovery, we do not believe that this will be the case this time. We believe recent more episodes (97198, 08/09) were reflective of more general market improvements and a retum to functioning capital markets (post the Russian default and post the financial crisis, respectively). These overall market recoveries, while important to oil, particularly the demand side, were somewhat outside direct, physical oil markets. Where this bear oil market is concerned, we don't see a quick macro event outside of physical oil markets as the panacea or quick fix. Rather, we are looking for a more "U" or "L" shaped recovery for oil. In 1986, again, oil priced quickly raced to the bottom over an 8-month period, but then, excluding the first Iraq war, took about 14 years to recover. To be clear, we are not making that extended recovery call here because we do believe that there were anomalies to the 1986 case. One anomaly being the 1986 Tax Reform Act and its net effect on overall US GDP, which according to the World Trade Organization, would normally have seen a greater positive effect from lower oil. However, even looking at the more "normalized" oil recoveries listed (excluding 1986 and 2009), it took oil anywhere from 11 to 26 months to return to its previous peak, leading to an average 17-month oil "recovery" time. Importantly, we acknowledge that markets are forward looking, and therefore believe that a change in sentiment itself could likely precede the actual improvement in oil, and also energy-related securities. We focus here on those data points that could improve the market mentality around oil. Note: we discuss the expected recovery cycle for high yield energy bonds specifically later in this outlook. Deutsche Bank Securities Inc. Page 5 EFTA01097659 13 January 2015 HY Corporate Credit Energy IFigure 4: Oil Supply-Demand Forecast 2013 2014e Nov-14 2015e 2016e 2014/13 2015(14 2016/15 2015/Nov-14 Consideratations OECD 46.1 45.6 45.6 45.5 -0.5 0.0 -0.1 US vs Europe GDP outlook Non-OECD 45.7 46.8 47.8 49.3 1.1 1.0 1.4 How resilient is China demand? Oil Demand 91.7 92.4 93.4 94.8 0.7 OS 1.3 Will lower oil price prompt upgrades? US 10.2 11.7 12.7 13.3 1.5 1.0 0.7 Pace & magnitude of supply response Other non.OPEC 44.4 44.8 45.1 45.5 0.4 0.3 0.6 Risk of delays, disruption & capex cuts Non-OPEC Supply 54.6 56.5 57.8 58.8 1.8 1.3 13 OPEC NGLa 6.3 SO 6.7 6.8 0.1 0.3 0.1 Libya 0.9 0.4 0.7 0.7 0.9 -0.5 0.3 0.0 0.0 Is there further downside supply? Iran 2.7 2.8 2.8 2.8 2.8 0.t 0 0.0 0.0 Possibility of mid-15 nuclear deal Iraq 3.1 3.3 3.4 3.4 3.3 0.2 0.1 0.0 0.0 Other 24.2 23.2 23.4 22.0 22.5 -t.0 -1.2 -0.3 -1.4 Will Saudi add supply if non-OPEC falls? Cell on OPEC 309 29.5 30.3 28.9 292 -1.4 -0.6 -0.3 -1.4 1.4mb/d reduction needed vs Nov-14 &wet De teary Sanwa, Looking at this simplified supply and demand outlook for oil sums up our thoughts on a longer term recovery. Given no expectation for a call on OPEC in 2015, we believe oil prices will persist at lower levels through 2015, especially in 1H 15 as the market is currently 1.4 million b/d oversupplied. The basic levers that could improve the supply and demand dynamics in the next year are the following in our minds: (i) better than expected economic growth (ie demand), (ii) more volatile seasonal weather patterns, (iii) faster than expected Non-OPEC production declines, (iv) a sooner-than-scheduled OPEC meeting (June currently), (iv) a sooner than scheduled OPEC meeting (June currently), or (v) unexpected geopolitically-related production declines. IFigure 5 GDP Forecast & Revision (% yoy) Forecast level Forecast change since Dec' 14 WO Sep' 14 WO' 2019E 2015E 2015E 2019E 2015E 2018E G7 1.8 2.5 2.4 -0.1 0.0 0.0 US 2.4 3.5 3.1 0.1 0.1 -01 Japan 0.5 1.4 1.6 -0.6 0.1 0.2 Euro area 0.8 1.0 1.3 0.0 -0.1 -0.1 Asia (ex-Japan) 6.0 6.2 6.1 -0.3 -0.7 -0.7 China 7.3 7.0 6.7 -0.5 -1.0 -1.3 India 5.5 6.5 6.5 0.0 0.0 0.0 EEMEA 2.3 1.9 2.5 0.4 -0.8 -0.4 Russia 0.5 -0.9 -0.4 0.0 -1.9 -1.8 Latin America 0.8 1.5 2.9 -0.2 -0.6 -01 Brazil 0.1 0.7 t.9 -0.2 -0.5 0.0 Advanced economies 1.7 2.4 2.3 0.0 0.0 0.0 EM economies 4.4 4.5 4.9 -0.2 -0.7 -0.6 Global 3.2 3.6 3.8 0.0 -0.3 -0.2 Sane Onift:Oe &Pat tco, tvwci Noe 'Sommer( OWN Onto* losakus Are torn ma%caned lamp Orrobei nfOPPP*arts Page 6 Deutsche Bank Securities Inc. EFTA01097660 13 January 2015 HY Corporate Credit Energy First, we look at the demand side. Looking at what our economics team is expecting for 2015, it is clear that despite the drop in oil prices (and its positive follow-on effects for certain economies), expected global growth trends have seen a notable decline over the last quarter of 2014. This is evidenced by the deceleration of expected growth reflected in our estimates, especially in Asia. The main drivers of the downgrades are driven "entirely by markdowns to emerging market growth prospects, which were reduced by more than 1/2 percentage points over the next two years. These downgrades were broad- based across regions, with three of the BRICs, Russia, China, and Brazil (in that order) recording the largest and most important downward revisions. Most important is the downward revision to China, where we now see increased negative spillover from past overinvestment in property and the government focused on a more sustainable 7% rate of growth. Global growth is expected to bottom at a relatively subdued rate of 3.2% in 2014 and rise slowly over the next two years." We see little possible upside here, and in fact, believe there could be further downside to global demand. Next, we look at the seasonality of oil and the possibility for more volatile seasonal weather patterns. Historical and expected oil demand seasonality is reflected in Figure 6 on the left. In the chart on the right, we overlay DB's quarterly demand/non-OPEC supply projections holding OPEC production flat at the October 2014 level (30.6 million b/d). Looking at Figure 7, one can see that the magnitude of the expected level of oversupply (1.5 million b/d in 1H 15) is well above historical averages. While many would look to Libya as the inevitable OPEC producer most likely to decrease production given recent unrest, even assuming no production from Libya, DB estimated excess oil supply in 1H 15 would still be 0.7 million b/d. This would leave the market needing a significant and unlikely change in weather to absorb excess supply in 1H 15. For example, last year a colder-than-normal winter increased seasonal demand by —250-300K b/d. On the margin, we see the seasonal demand effects of oil as well as marginal economics (discussed below) as a key reason for sub-S50/bbl oil at least through 1H 15. Figure 6:1H Global Crude Demand Typically 1 Million b/d Figure 7:2015 Will Require Market to Absorb >Normal less than FY Seasonal Excess LON ton counter seasonal move In 1)00 stocks/other m 2H14 to absorb over-supply 1.000 1400 Seasonality typically means that III demand is —1mb/d 000 140 below the FY avg. 0 400 400 4,500 If we assume NOreduction InOPEC supply Vs 4014 pressue to absorb 4400 4.000 peaks In 2Q15 with larger than normal IH stock build required -1400 4.500 01 02 CO 01 Oi a to ea • 2011 *2012 •201.1 420W0 42010. •201I ann 52013 salll. noise Sauces LA DitEtonorna D (ecotone* Sam* Moving onto supply, we look at Non-OPEC production. For purposes of this discussion, we are talking largely about US production growth as markets (and OPEC) are now looking towards the US for a supply response (i.e. significant reduction in annual oil growth). However, we do not think that supply response will be one readily observed in 2015 largely due to (i) a significant inventory of drilled but uncompleted wells (up to 6 months of inventory per producer), (ii) a relatively high level of hedging in 2015 (less so in 2016), (iii) the ability to high grade to the most economic plays, (iv) recently achieved drilling Deutsche Bank Securities Inc. Page 7 EFTA01097661 13 January 2015 HY Corporate Credit Energy efficiency, and (v) decreasing non-productive capital spending (test wells, seismic, infrastructure, etc). As we consider 2016, our equity counterparts have looked at the issue of US growth and asked the question: "What would we need to see from the industry to normalize production growth from 2016 forward at a level more consistent with demand expectations? (i.e. reduce US YoY supply growth to 500-600 Mb/d)." They estimate "that at minimum, the industry would need to drop —160+ horizontal rigs from the "Big Three" plays (Bakken, Permian, Eagle Ford) - or -25% of the near 700 rigs currently operating in these plays. The implications for the overall oil rig count (-1,600) is much more severe, as vertical rigs and marginal plays would likely fall off first, implying a total rig cut of 500+." If this scenario plays out as expected, it suggests that US YoY supply growth would be reduced from 0.90.1.0 million boepd now to about 500-600K boepd, or a reduction of 300-400K boepd. We believe markets would view this move as significant. Looking at Figure 8, the good news is that we have seen some solid progress towards decreasing horizontal rigs by that —160 figure. Since the OPEC meeting in late November, horizontal rig count in the Big Three plays has decreased by 40 rigs or about 25% of the DB required cut - and the biggest step down came in this past week. In that same span, total US rig count has decreased by 167 rigs or about 33% of the DB required cut - again, with the best step down coming this past week. Not surprisingly, the necessary decline in less efficient rigs (both vertical rigs and rigs in marginal plays) is happening faster than for the core three plays, where rigs are most efficient. IFigure 8: US Drilling Rig Score Card Since November OPEC Meeting Total Rigs Across All Basins Directional Horizontal Vertical Total Rigs working as of OPEC Meeting 194 1,371 352 1,917 12/5/2014 WoW change 4 (3) 2 3 12/12)2014 WoW change 121 (II (24) 127) 12719/2014 WoW Mange Ill 111l (6) 118) 12/26/2014 WoW Mange (141 161 (15) 135) 1/2)2015 wow change (61 (141 19) 129) 1/9/2015 WoW change (141 (351 (12) 161) Rigs working as Week 1/9 161 1,301 288 1,750 Total Rig Decline since OPEC Meeting (11/27) (33) (70) (64) (167) Total Rig Decline since Relative Oil Peak 16/20) (671 51 (92) (108) Horizinal Rig Decline in the Big 3 Plays Eagle Ford Permian Bakken Total Riga working ea of OPEC Meeting 207 363 189 769 12/5/2014 WoW change 131 4 (3) (2) 12712/2014 WoW change 121 171 (1) 110) 12/19/2014 WoW change 2 4 (7) (1) 12/26/2014 WoW change 121 3 (2) (1) 1/2/2015 WoW change (51 0 0 (5) 1/9/2015 WoW change 131 (101 (8) 121) Rigs working as Week 1/9 194 357 168 719 Big 3 Rig Decline since OPEC Meeting (11/27) 1131 (61 (21) (40) Big 3 Rig Decline since Relative Oil Peak (6/20) 10 (34) 7 117) Savo. Bab/ I& Page 8 Deutsche Bank Securities Inc. EFTA01097662 13 January 2015 HY Corporate Credit Energy Beyond these initially positive data points on lower rig count, there are upside supply risks in the form of high grading and drilling innovation. The natural gas playbook has shown us that producers across the board have proven to be innovative during price dislocations like the one we are currently experiencing. Over the past several years, the path of continued drilling improvement has been the norm for producers. While some plays are further along the innovation road (Bakken, Eagle Ford) than others, we believe the current environment will only incentivize producers to push for further process improvements; we should continue to see more production with less capital. That said, as producers narrow their focus and look to high grade to only the top plays (Permian), other relatively less attractive plays like the Bakken and Eagle Ford will see drilling dollars move out. One can see this above where Permian horizontal rig count has gone down by just 6 rigs since the OPEC meeting, and Bakken and Eagle Ford horizontal rig count has declined by 21 and 13 rigs, respectively. The offset to the high grading and efficiency points would be an inability to regain momentum. Producers have also done an excellent job maintaining momentum over the past several years by securing low cost capital, protecting cash flows with hedges and innovating operationally; prolonged, depressed oil prices will eliminate two of those levers for a time, limiting how and where E&Ps can allocate spending dollars most efficiently to generate an acceptable return. Moving on to OPEC, when looking at the recent non-action with our energy hat on, it clearly shows OPEC, and really Saudi Arabia's, intention to maintain market share in the long term. It plainly had two decisions - either let higher cost Non-OPEC oil growth continue to encroach and likely accelerate over time, or take a stand to undermine Non-OPEC supply now before it became a bigger threat. The OPEC decision does indeed make sense on a stand-alone basis, but there are a couple other "ancillary" benefits to OPEC/Saudi Arabia including inflicting pain directly on both Iran and Russia through lower export revenues. Pressuring Russia further benefits Saudi Arabia as Russia will be a weaker supporter of Iran, its most significant rival in the region. Given that Saudi Arabia has about twice the level of govemment assets Russia does (on an annual basis: government assets divided by budget deficit), it can afford to withstand a low-price oil environment for longer to protect its long-term market share. 1 Figure 9: US Onshore Oil Breakeven Economics ▪ 00 • 410 • $(0421 $41.170 L ON - MOS VS SO) • 20 US 350) • S32$90 •1%599 3>1% 3," ?MO ?MO a 1270 110 5,0 i 2014 mr in* Sweet Wood illakania; DB Equey Rtrakarch Wood Mackenzie has done analysis to show the medium to long term breakeven economics of US unconventional crude supply (above). It clearly shows that a majority of US unconventional plays break even in the S65-70/bbl Deutsche Bank Securities Inc. Page 9 EFTA01097663 13 January 2015 HY Corporate Credit Energy area. It is important to keep those levels in mind when considering OPEC's next move. In order to inflict maximum damage to base US production, oil prices need to stay below long term breakeven levels for a sustained period of time. On the flip side, US E&Ps will do what they can in the short term to side- step a permanent momentum shift in their core plays. As we have seen, oil prices are slowly converging on this short-term, US unconventional "marginal" cost ($15-30/bbl range, discussed later). Taking all this together, we do think OPEC is prepared to maintain current supply at this level to maintain share in the long term. If it stopped short of its intended goal of slowing US supply, why embark on this path in the first place? Lastly we consider the possibility geopolitically-related and other production decreases or increases. Libya comes to mind first as recent unrest shutdown its El Sharara field, which was producing 270K b/d. This decreased November production down to 638K b/d from its previous peak of 883K b/d. As lower oil prices continue to pressure Venezuela and its stability, that will be another country to watch. Rounding out the areas of interest are West African deepwater projects and Russian Arctic projects - both of which are on the high-end of the breakeven cost spectrum. Beyond these specific areas to watch, we would point out Non-OPEC related growth does tend to underwhelm, and it would need to outperform on a larger scale in 2015 to offer any meaningful upside surprise large enough to affect oil prices. The last wild card we would point out here is China and its Strategic Petroleum Reserve (SPR). China's SPR purchases averaged —500K b/d for most of 2014. Now, Chinese SPR-related imports are thought to have peaked in December as the country took advantage of depressed oil prices with upwards of 700K b/d in estimated purchases. Despite brisk Chinese demand last month, most expect this specific demand level to moderate moving forward as the country is apparently close to filling existing reserve capacity. The combination of all of the previous five factors, seasonality in particular, is clearly reflected in investors current thinking on oil prices. DB recently surveyed equity investors with a majority of those surveyed (70%) of the belief that oil will bottom by the end of 1H 15. In particular, that same set of investors overwhelmingly believes that North American crude supply is the biggest driver of oil prices (50% ranked it the top factor). We would tend to agree more with the second observation and believe that the signs of slowing North American production are key to the recovery in oil prices after 1H 15 seasonality plays through; we believe improving oil price will require some positive oil supply data points (e.g. inventory draws coming in 2H 15, continued progress on decreasing horizontal rigs in the Big Three oil plays) to give that psychological boost needed to change oil market momentum, even if those positive data points don't immediately translate to lower observed supply growth. In the near term (1H 15), we would expect to see sub-$50 oil persist as the market continues to be oversupplied by1.4 million b/d. Also, to put current trading levels into perspective, WTI prices would need to fall to $45 to surpass the 58% collapse in the oil price during the Jan 97/Dec 98 episode; it seems appropriate to compare recent price action to that episode since during the first half of this year global oil supply will be growing in excess of global oil demand by its largest margin since 1998. Specifically, we think this 97/98 case also speaks to the possibility of 2015 being a lost year, and thus, investors focus on 2016 as the next meaningful point in time for price recovery. Given we are approaching what could be the lows in oil, where we with HY E&P in the credit cycle now? While we have seen many episodes where oil prices have declined over 50%, none of those were prolonged enough to coincide with a restructuring of the entire E&P sector during a time when we had a fully developed and functioning HY market. For example, the 1986 episode did illustrate a Page 10 Deutsche Bank Securities Inc. EFTA01097664 13 January 2015 HY Corporate Credit Energy prolonged bear market in oil, but there is no reliable data for that period. What this table and particular analysis shows us is that we are at and beyond the average decline for an oil cycle, which is a 50% decline. That said, while finding the bottom for oil is important, the recovery story is equally important. IFigure 10: Cumulative cyclical peak default rates in HY and IRRs on energy bonds from current levels Year 1 Year 2 Year 3 To: Year 1 Year 2 Year 3 Cyclical Peak Cumulative Default Rates Coupon Price Surviving Par BBB 0.5 1.1 1.8 5.5 106.0 BBB 100 100 99 99 BB 1.7 6.6 10.1 6.0 95.8 13B 100 99 96 94 B 5.7 18.2 25.1 7.4 81.9 8 100 97 91 87 CCC 19.9 42.3 51.6 8.4 59.2 CCC 100 94 86 83 All HI 5.7 16.7 22.9 6.8 85.5 HY 100 97 91 88 Single-Bs/CCCs 32.4 Coupon BBB 5.5 5.5 5.4 BB 5.9 5.8 5.7 8 7.3 6.9 6.6 CCC 8.1 7.6 7.1 HY 6.7 6.4 6.1 Percent Downgraded BBB 20 10 5 BB 25 13 6 8 .- _ _ CCC — Dollar Prices, Adjusted for Downgrades BBB 106 102 101 100 BB 96 89 87 86 a 82 82 82 82 CCC 59 59 59 59 HY 86 82 81 81 Market Value (Credit Loss . Coupon 4 Downgrades) IRR BBB 106 107 106 105 -1.2 BB 96 94 90 87 -9.4 B 82 87 81 78 -4.8 CCC 59 64 59 56 -4.7 HY 86 87 82 79 -7.3 Sousa °anent Ban*Ove4 arraroev While we don't have a specific example of an energy sector restructuring, our counterparts in credit strategy have done work on cumulative cycle peak default rates in the overall HY sector. In Figure 10. they show average cumulative default rates in HY over the last three credit cycles (measured in 1989-1991, 2000.2002, and 2008-2010 by Moody's issuer-weighted rates). According to this data, we are looking at an average cumulative default rate of 22.9% for the whole HY spectrum and 51.6% for CCCs only. Looking at the bottom section, titled Market Value, $59 invested in an average CCC energy bond today should return $56 at the end of year 3 yielding a -4.7% IRR. Percent changes on the right are calculated between these two values, non- annualized. So how do we interpret these numbers? The good news here is that using a relatively negative set of assumptions — cyclical peak default and downgrade rates (based on broad market historical stets) and unchanged dollar prices on the exit from year three — are resulting in only single-digit negative retums from here, implying that to a good extent, the bad news has been priced in. An important bullish assumption that we are purposefully not including here - that a surviving CCC could be worth more than $59 at the end Deutsche Bank Securities Inc. Page 11 EFTA01097665 13 January 2015 HY Corporate Credit Energy of year 3 - could single-handedly result in positive IRRs. Consider: only three months after the end previous two credit cycles in 2002 and 2009, an average CCC jumped in price from low 50-ies to low 70-ies (measured in June 2003 and June 2009). If we were to plug in a $70 year 3 price assumption for surviving CCCs in Figure 10, non-annualized IRR jumps to +10%. For a more detailed explanation of assumptions to reach these conclusions, please see DB's US Credit Strategy: What is Priced in Energy Bonds Here? (December 18, 2014). Despite a dearth of examples around a crisis in HY energy, or more specifically Figure 11: Widest Levels of Each the E&P sub-sector, we do have some examples of where other distressed HY Crisis sectors traded at crisis levels (right). If these restructured sectors hold any 88s 8s CCCs weight, it seems by this metric, we still have downside on a spread-basis Real Estate 12/31/2008 1,573 2,008 4,702 before we reach the bottom for HY E&P bonds. Despite somewhat conflicting answers from the above two analyses, we think the takeaway for HY credit is Media 11/30/2008 1,128 2,029 3,508 that we are getting there but not be at the optimal entry point for bonds. Autos 12/31/2008 1,546 2,036 2,473 Further, investors need to consider time horizons as they relate to various Telecoms 07/31/2002 1,398 1.014 3,966 investing strategies and whether or not funds are locked-in. Lastly, with a Gaming 11/30/2008 1,895 1,870 2,485 significant level of default (blended rate of 30% for B/CCC), there will be clear Average 1,508 1,791 3,427 winners and losers, outcomes will be very binary outside of buying a diversified basket of lower-quality HY E&P credits. Energy 01/08/2015 465 960 1,896 % of other crisis levels 30.9%53.6% 55.3% What can we expect in terms default rates and recovery should there be a full- Sane Dooncft Bent scale restructuring of energy names? As we have not seen a full scale restructuring of the sector is hard to pinpoint exact figures here. Moodys reported that during the last two credit cycles for E&Ps (1998-1999, 2001-2002), a majority of unsecured creditors received a range of 35-50% recoveries with an average recovery of 40%. This is better than the average industrial recovery of 29%. Fitch has reported similar numbers. From 2000-2013, the average recovery rate for energy was 45%; compared to a 37% for the total market. However, the spread on recoveries in any given year was relatively wide with a low of 8% in 2001 and a high of 76% in 2011 looking at the Fitch data. Important to note that per Fitch, the comparative default rates for energy have been relatively mild at 2.0% from 1980-2012, this compares to a 4.6% default rate for the overall market during the same period. What other major catalysts need to play out to trigger defaults for HY E&Ps? As we look at maturities due in HY energy over the next couple of years (Figure 12), it is clear that most energy companies including E&Ps have termed out debt. This is not surprising given the historically low rates issuers have achieved in the HY market over the last several years. One can see that there is a little less than $4 billion in overall energy HY bonds coming due in the next two yea