E&P & Refiners. Permian Crude Takeaway Constraints: Winners and Losers. 13 April 2018 Americas/United States Equity Research Oil & Gas

Transcription

1 Americas/United States Equity Research Oil & Gas Research Analysts William Featherston Betty Jiang, CFA Manav Gupta William Janela, CFA Michael Ziffer, CFA Chris Baker, CFA Christopher Zhang, CFA E&P & Refiners THEME Permian Crude Takeaway Constraints: Winners and Losers With Permian diffs already widening and the next large slug of takeaway capacity >12 months away, we revisit Permian Basin supply/demand balances, widen our base case Midland differentials through 2019, and take a closer look at which companies are best/worst positioned for likely further widening in basis. With ~3.6 MMBbld of current demand (3.2 MMBbld pipeline MBbld refineries) vs. ~3.5 MMBbld of supply estimated at YE18 and 3.8 MMBbld by 2Q19, production is already above 84% of takeaway utilization and poised to exceed 95% from 4Q18-2Q19 before the next leg of pipeline expansions provides relief in 2H19. Given supply outgrows capacity between 3Q18 and 2Q19, we expect Midland to Cushing spreads will widen from breakeven in 2017 to $8/Bbl by YE18 and remain wide until two large pipelines come in-service in 2H19. New pipeline capacity. With EPD s 200 MBbld Midland to Sealy pipeline having just reached capacity, the next large chunk of incremental takeaway capacity is the 585 MBbld Cactus II, scheduled to come online in 3Q19. Permian production outlook. Between now and 2Q19, we forecast Permian production grows from 3.1 MMBbld to 3.8 MMBbld, exceeding takeaway capacity and forcing producers to employ high cost trucking to move barrels, widening basis materially, and risking production shortfalls. Separating the haves from have-nots. Permian E&Ps that control gathering from the wellhead to the pipeline inlet, and have firm transportation (FT) pipeline agreements to higher price Gulf Coast markets are best positioned to avoid Permian congestion and wide Midland basis. This report details FT arrangements and percentage exposure to Midland pricing for each Permian-leveraged E&P in our universe on page Stock thoughts. Wider Mid-Cush differentials are a positive for Permianlevered refiners. A more tempered Permian production growth outlook is also positive for oil macro fundamentals, benefitting non-permian oil-weighted E&Ps (Bakken in particular), and Permian E&Ps with FT out of the basin or pricing contracts off of Gulf Coast/Houston benchmarks. Large-cap E&Ps: OXY is the best positioned E&P benefitting from widening basis, while PXD is very well positioned with >90% FT to the Gulf Coast. Permian-levered SMID-cap E&Ps: Most of WPX s volumes are covered by FT to Corpus Christi, while LPI, PE, and QEP have pricing agreements off of Gulf Coast and/or Cushing markets on a majority of their production. Refiners. Within refiners, DK, HFC and ANDV will be the beneficiaries of wider diffs. We raise DK PT to $54, HFC s PT to $57, and ANDV PT to $142. DISCLOSURE APPENDIX AT THE BACK OF THIS REPORT CONTAINS IMPORTANT DISCLOSURES, ANALYST CERTIFICATIONS, LEGAL ENTITY DISCLOSURE AND THE STATUS OF NON-US ANALYSTS. US Disclosure: Credit Suisse does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the Firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.

2 Contents Contents 2 Introduction 3 Implications to E&Ps 5 Best Strategic Positions 5 Best Positioned E&Ps 6 Greater Pricing Exposure 7 Implication for Refiners 12 Permian Supply/Demand Considerations 14 Where we stand on production 14 Near-term milestone / catalyst: What to watch over the next 18 months? 14 How wide can Mid-Cush differential get? 15 Greenfield Projects 18 Midland-to-Sealy Pipeline (EPD)...19 EPIC Pipeline (ARES Capital)...20 Cactus II (PAA)...21 Grey Oak Pipeline (PSX / ENB)...22 Permian to Corpus / Houston (MMP)...23 Permian-to-Nederland (ETP)...24 South Texas Gateway (BPL)...25 Large Brownfield Expansions / Conversions 26 Permian Express III Phase 2 (ETP)...27 Sunrise Pipeline Extension + Loop (PAA)...28 NGL Pipeline Conversion (EPD)...29 Select Existing Pipelines 30 Basin (PAA)...31 Centurion (OXY)...32 Cactus I (PAA)...33 Longhorn (MMP)...34 BridgeTex (MMP / PAA)...35 West Texas Gulf (ETP)...36 Permian Express I and II (ETP)...37 PELA (ETP)...38 E&P & Refiners 2

3 Introduction Given the debate around a potential blowout in Permian diffs, we have revisited our Permian Basin supply/demand balances and take a closer look at new crude takeaway projects that would alleviate potential bottlenecks. With ~3.64 MMBbld of current demand (3.19 MMBbld pipeline MBbld refineries) vs. ~3.5 MMBbld of supply estimated at YE18, pipeline takeaway is clearly getting tighter as Permian production grows. Futures are indicating Midland trading at a ~$5.50/Bbl discount to WTI by YE18. For context, we would highlight that a month back the spread was only ~$3/bbl and 6 months back it was at transportation cost of ~$0.75/bbl. Moreover, the current spread of $4.80/Bbl is likely to widen from here given expectations of >700 MBbld of production growth before the next big pipeline takeaway project comes online. Figure 1: Midland -Cushing Spread ($/bbl) Figure 2: Futures Midland Cushing Spread ($/bbl) 5.00 $ Jan-11 Jun-11 Nov-11 Apr-12 Sep-12 Feb-13 Jul-13 Dec-13 May-14 Oct-14 Mar-15 Aug-15 Jan-16 Jun-16 Nov-16 Apr-17 Sep-17 Feb-18 $0.00 -$1.00 -$2.00 -$ $4.00 -$ $ Midland - Cushing (today) Midland - Cushing (6 Month ago) Midland - Cushing (1 Month ago) Midland - Cushing (1 year ago) Source: the BLOOMBERG PROFESSIONAL service Source: the BLOOMBERG PROFESSIONAL service We estimate Permian Basin takeaway capacity currently sits at 3.64 MMBbld (3.192 MMBbld on pipeline and 448 MBbld of local refining capacity) following the recent expansion of Enterprise Product Partner s Midland to Sealy 200 MBbld expansion. With 2Q Permian production of 3.1 MMBbld, takeaway capacity utilization currently is ~84% and we generally see basis widening when utilization rates exceed 85%. With the differential widening already signaling tightness in the physical market, the situation is likely to get much worse in late 2018 with continued Permian growth and no incremental takeaway capacity coming online until 1Q19 (Sunrise Expansion s 120 MBbld) and nothing meaningful until the 585 MBbld Cactus II pipeline comes online in 3Q19. For perspective, we estimate Permian production to add another 325 MBbld from 2Q18 to 4Q18, pushing takeaway capacity utilization to 92% by YE18. And in 1H19, we see another 400 MBbld growth by 2Q19, pushing the capacity utilization to nearly 100% and forcing operators to the last resort of higher cost trucking of volumes resulting in a further widening in basis. Solution is 18 months away. By 4Q19, the combination of the 585 MBbld Cactus II expansion and the 590 MBbld EPIC Pipeline should materially boost Permian Basin pipeline takeaway capacity to 5.2 MMBbld. With production forecasted to be 4.2 MMBbld, takeaway capacity utilization should decline appreciably to ~80% by YE18 and we forecast a further decline to <80% in 2020 as another wave of pipeline expansions and greenfield projects are scheduled to come online. See Figure 20 for detailed project buildout schedule. E&P & Refiners 3

4 Figure 3: CSe Permian Crude Production vs. Takeaway Capacity ( ) CS Permian Oil Production Forecast with % Capacity Utilization MMBbld % 83% 87% 92% 93% 98% 86% 80% Expected Expansion Capacity 2.0 Current Takeaway Capacity 1.0 Rail 0.0 Local Refinery Capacity Source: Credit Suisse estimates Our bear case assumes the Midland-Cushing spread blows out to $12/Bbl in 4Q18 and 1Q19, ~$4/Bbl above our base case. We think the following factors could drive spreads wider than our base case 1) Planned / unplanned refinery downtime in the Permian; 2) logistics logjam, increasing trucking demand in particular; 3) Delays that push back the in service date of pipelines; 4) Pipelines refusing to carry super light crude that fails to meet pipeline specs; 5) Permian production exceeding our estimations. We believe, any one or a combination of these factors could result in crude glut filling storage at a faster pace than expected, leaving producers with the only option of railing crudes to either coasts (East or West) that would cost them between $12-$14/bbl. Figure 4: Mid-Cush price deck scenarios and strip $1 $(0.50) ($1) ($1) Mid-Cush Diff ($/bbl) ($3) ($5) ($7) ($9) ($5) ($6) ($8) ($8) ($6) ($2) ($11) Bull/Bear Range Base Case ($13) Current Strip Source: Credit Suisse estimates E&P & Refiners 4

5 Implications to E&Ps Best Strategic Positions E&P marketing strategies vary greatly, but the best positioned Permian E&Ps are those that control the full marketing value chain and have own firm transportation agreements to higher priced more flexible Gulf Coast markets. Companies trucking barrels to local west Texas markets are most disadvantaged, receiving lower prices and at greater risk of logistical bottlenecks causing delays to growth plans. Before we dive into company level strategy, it s worthwhile to give a quick overview of the oil marketing value chain. Broadly speaking, oil volumes produced are stored on the lease in tank batteries, which are moved off lease through either company-owned or 3 rd party midstream pipeline gathering system or via trucks. Oil is then transported to central delivery points that feed into long haul pipelines that move volumes out of basin. For most operators, oil is sold at the wellhead, meaning transfer of custody happens at the lease level with 3 rd party midstream marketers controlling actual sale and transport of barrels out of basin. In those cases, the netback to the upstream operator varies by specific contract, but is generally priced off of the local Midland benchmark. However, if the pipelines are at or near full capacity, for production exceeds capacity available from 3 rd party midstream providers, it s unclear how midstream would allocate its pipeline capacity among its customers. As such, we d generally categorize exposure into 3 broader groups: 1) Companies with firm takeaway or sale agreements that provides flow assurance AND pricing advantage to Gulf Coast/Houston markets. 2) Companies with trucked barrels without pipeline sales agreements that would need to go farther to find pipeline injection point with available capacity, or rail delivery point at a higher cost, or ultimately truck to end-market locations. 3) Everything else in between that are sold at the wellhead, and priced off of Midland. E&Ps without firm agreements out of basin and instead rely on 3 rd party midstream, face more pricing uncertainty when pipelines reach full capacity. Our analysis show that Large-Caps are in general are much better positioned than SMID-cap E&Ps. Companies best positioned are those that control their destiny from wellhead to end-market by using company-owned gathering system, and have firm takeaway (FT) on long-haul pipelines to move crude to more favorably priced destinations i.e. East Houston and Corpus Christi. We outline oil marketing strategy by company on page E&P & Refiners 5

6 Figure 5: E&P Permian oil marketing relative positioning (before-hedges) Better pricing OXY, PXD*, LPI*, APC, WPX Gathers & markets most of their own oil volumes / Predominantly priced off of Gulf Coast/Houston benchmark prices Greater Operational Control PE, EOG, DVN, QEP NBL, APA FANG, CDEV, CPE CXO, EGN, SM, XEC Markets most of their own oil volumes / Some exposure to prices in Gulf Coast/Houston and/or Cushing Sell at the wellhead / Midland pricing / Have contracts with 3rd party midstream providing flow assurance out of basin Sell at the wellhead / Midland pricing / Have gathering agreements with FT or contracts to Midland Sell at the wellhead / Midland pricing / Third party gathering PDCE Trucking / Midland pricing exposure Source: Company data, Credit Suisse estimates *Grouped with peers which have similar pricing exposure, PXD, and LPI use third party gathering. Grouped with peers which have similar pricing exposure, QEP sells at the wellhead but have 75% volumes indexed to non-midland prices Best Positioned E&Ps Occidental Petroleum (Neutral) is arguably the best positioned E&P company as it should actually benefit materially from a widening in basis due to their marketing arm, Occidental Energy Marketing Inc. (OEMI). With 1Q18E Permian production of 224 MBbld expected to ramp to 277 MBbld by 4Q19, OXY has 600 MBbld of firm transportation (FT) on four pipelines that deliver to Houston and the Gulf Coast: Cactus-200 MBbld; BridgeTex-200 MBbld; Longhorn-50 MBbld; and Sunoco-20 MBbld. OXY will further increase its FT when Cactus II comes on-stream in 3Q19. Most of this FT is held through 2025 and enables OXY to deliver its ~250 MBbld of production is well as enable its marketing arm to purchase volumes in Midland from other E&Ps and deliver it to the Gulf Coast/Houston at higher prices and collect the margin. As OXY s upstream business sells its crude internally to OEMI at Midland prices, we estimate every $1/Bbl widening in basis hurts its upstream cash flow by $90 million annually. However, that is more than offset by the $180 million benefit OEMI realizes with every $1/Bbl widening in basis. In aggregate, OXY should see a $90 million per annum benefit (0r $0.12/share) for each $1/Bbl widening in basis. On the natural gas side, 75% of OXY s production is delivered on Oasis with FT contracts to Katy, TX reducing its exposure to the widening Waha-Henry Hub basis blowout. Pioneer Natural Resources (Outperform) is the second best positioned E&P company with regards to widening Midland basis. With over 160 MBbld of FT capacity on Cactus, Magellan, BridgeTex and Permian Express, PXD has over 90% of its 2018 production being delivered to either Houston or the Gulf Coast for export. So while many Permian E&Ps are likely to post lower realizations in 2Q18-2Q19 because of widening basis, PXD s realizations will continue to be leveraged to Houston/Gulf Coast and export markets less the cost of its FT of ~$3/Bbl. As PXD is expected to grow its Permian production >20% per annum, its FT will increase and is expected to cover 70-80% of its volumes insulating it from widening basis. PXD also has 233 MMcfd of FT on EPNG to southern California covering virtually all of their estimated 2018 gas production. But as gas volumes ramp in 2019, we estimate 25% of their gas volumes will be exposed to Waha basis in 1H19 until Gulf Coast Express (where PXD has contracted for FT) comes on stream in 3Q19. E&P & Refiners 6

7 WPX Energy (Outperform) is differentiated on its takeaway strategy being the only SMID-cap E&P that gathers and markets its own oil and gas in the Permian. On oil, WPX flows through its 50% owned crude gathering system where current capacity is 125 MBbld, and it has 100 MBbld of gathering contract with Oryx II, which comes online in Nov 18. In 2018, WPX has ~30 MBbld of FT to Corpus Christi and 5 MBbld of FT from Midland to Cushing, which doubles to MBbld to Corpus Christi in The company also has 14.5/20 MBbld of Mid-Cush basis hedges in 2018/19 at ~($0.90)/Bbl. The combined basis protection from FT and hedges exceed our Permian oil production forecast, meaning WPX will actually have a net benefit to wider Midland differential over the next 2 years. Similarly on natural gas, WPX is also well positioned holding up to 500 MMBtu/d FT on the WhiteWater pipeline from their Stateline asset to Waha (expected in service in 1H18) and another 200 MMBtu/d FT on Atmos pipeline taking gas out of Waha to Katy, TX (Houston). Both oil and gas agreements are well in excess of its production, providing both uplifted pricing and flow assurance. Laredo Petroleum (Neutral) is well positioned on oil takeaway with 80% of oil production gathered on pipe on company-owned pipelines, which is connected to the larger Medallion gathering system that provides access to long haul pipelines. LPI has 14 MBbld net (19 MBbld gross) of firm sales to Shell through June 2020 and 10 MBbld net of FT on BridgeTex through 1Q25, both at Gulf Coast pricing. For the remaining volumes, it s protected by 10 MBbld of Mid-Cush hedges in 2018 at ~($0.56)/Bbl. In sum, their oil volumes are ~90% exposed to Gulf Coast pricing and would have a net benefit to wider differential due to its basis hedges. Even with no basis hedges in 2019, current agreements would only leave ~26% of our forecasted 2019 crude volumes exposed to Midland pricing. Parsley Energy (Outperform) is a co-anchor on Medallion s gathering system which connects to both Longhorn and BridgeTex longhaul pipelines. PE has firm sales contracts with several marketers for a total of 95 MBbld gross operated volumes in 2018/19 (only ~44 MBbld of which is take-or-pay) with pricing exposure evenly split ~33% Gulf Coast, ~33% Cushing, and ~33% Midland. Gulf Coast pricing exposure is fixed off of Magellan East Houston (MEH). Of the remaining barrels exposed to Midland pricing, they have ~11.4 MBbld of Mid-Cush hedges at ~$(0.86)/Bbl in 2018 which means only 25% of total net volumes are exposed to the Midland differential. Greater Pricing Exposure PDC Energy (Neutral) is still in the early stages of Delaware development and is currently trucking all of its volumes to central delivery points, which leaves them more exposed to potential takeaway constraints. While the pricing risk is protected by basis hedges, PDCE has higher risk of flow assurance if pipelines are at capacity, and volumes would need to be instead trucked on put on rail to Gulf Coast at a cost of $8- $10/Bbl. All that said, PDCE is building out their own oil gathering system for their Block 4 asset on the Eastern portion of their Delaware position in 2H18. And Delaware volumes only make up ~23% of total forecasted crude production in 2018, and a $1 widening in basis (versus our base case) only amounts to a ~$1.8MM impact to 2018 discretionary cash flow. The majority of SMID-cap E&Ps sell crude at the wellhead and are exposed to Midland pricing that is at risk of further weakness if more basin volumes are moved on higher cost alternative transportation options. Operators in this category can mitigate their cash flow impact via basis hedges. On a % of cash flow impact basis (Figure 6), companies with the most downside exposure are CDEV and FANG, due to their 100% Midland pricing exposure and limited basis hedges in , but both are well situated with already strong balance sheet and spending near cash flow. E&P & Refiners 7

8 Figure 6: Midland Pricing Exposure and Cash Flow Sensitivity For Permian-Levered E&Ps % Exposed to Midland Pricing (Pre-Hedge) Mid-Cush Basis Hedges % of Total Permian Oil Net Exposure to Midland Pricing (Post-Hedge) % Cash Flow Impact Bear vs Base Case Ticker CDEV 100% 100% 20% 0% 80% 100% -3.6% -5.3% -$8.5 -$17.8 FANG 100% 100% 17% 0% 83% 100% -3.5% -4.9% -$19.0 -$39.7 XEC 100% 100% 23% 6% 77% 94% -2.1% -3.3% -$11.6 -$24.5 EGN 100% 100% 52% 21% 48% 79% -2.0% -4.0% -$7.2 -$21.2 CXO 100% 100% 66% 30% 34% 70% -1.9% -3.5% -$15.1 -$59.7 CPE 100% 100% 60% 0% 40% 100% -1.7% -4.8% -$2.6 -$10.9 JAG 100% 100% 63% 24% 37% 76% -1.6% -4.0% -$2.3 -$8.4 SM 100% 100% 71% 28% 29% 72% -1.1% -3.5% -$2.8 -$14.3 PE ~42% ~56% 17% 0% 25% 56% -0.7% -1.6% -$2.9 -$10.1 PDCE 100% 100% 45% 0% 55% 100% -0.6% -1.3% -$1.8 -$6.0 WPX ~17% ~0% 34% 39% -17% -39% -0.4% 0.8% -$1.5 $4.7 LPI ~10% ~26% 34% 0% -24% 26% 0.5% -1.4% $1.2 -$3.6 QEP ~25% ~25% 74% 31% -49% -6% 1.4% 0.3% $4.2 $1.3 Average 76% 77% 44% 14% 32% 64% -1.3% -2.8% -$5.4 -$16.2 C Abs Cash Flow Impact to -$1/Bbl Basis Chg. ($MM) Source: Company data, Credit Suisse estimates Diversified Large-Cap E&Ps Relatively Marginal Exposure to Wider Basis Overall, the global large-cap E&Ps have relatively modest exposure to widening Midland crude differentials, as most have firm transportation agreements in place (or have committed capacity on future pipelines/expansions). And the Permian represents a much smaller portion of companywide oil production relative to pure-play operators in the basin. In Figure 7 below, we provide our estimates for Permian crude oil and Midland pricing exposure for the diversified large-caps with meaningful near-term production in the basin, ranked by estimated unhedged Permian oil production exposed to Midland pricing as a percentage of companywide oil volumes. DVN appears very well positioned into a widening Midland-Cushing spread as half of its production is delivered to Houston on a multi-year FT agreement on Longhorn, and the balance of its production is hedged at an attractive basis differential of ~$1/Bbl in both EOG is well diversified in terms of 3-5 year firm transportation agreements to the Gulf Coast (~20% of Permian production) and Cushing (>30% of production), with less than half of its Permian crude volumes levered to Midland pricing (we estimate ~35%, although it has not disclosed this figure). Moreover, of those barrels, it has hedged MBbld in at a basis differential of ~$1/Bbl, leaving just ~6% of companywide crude oil production exposed to widening Midland pricing differentials. We estimate just ~8-11% of APC s companywide E oil volumes are fully exposed to Midland pricing. The company currently has FT to move roughly half of its Permian crude to Gulf Coast markets, although it has agreements in place such that by YE19 it will have ~100% of basin oil volumes covered by FT. NBL s Permian oil volumes are currently sold at Midland pricing (Crane, TX), though it has secured up to ~75 MBbld of capacity on the ~590 MBbld EPIC pipeline (via acreage dedications) which is expected to start-up in late 2019, implying >50% of E Permian crude production will be sold into Gulf Coast markets. APA has contracted with third parties that own FT on various pipelines, providing it flow assurance to move crude out of the basin. However, it will be exposed to Midland pricing for all of its Permian oil (~35-40% of E companywide oil volumes). E&P & Refiners 8

9 Figure 7: Permian Oil Production and Midland Pricing Exposure for Diversified Large-Cap E&Ps Permian Oil (% of total oil volumes) % of Permian Oil Exposed to Midland Unhedged Midland-Exposed Bbls (% of total oil volumes) DVN 15% 17% 50% 50% 0% 0% EOG 28% 30% 35% 35% 6% 6% APC 17% 23% 50% 50% 8% 11% NBL 31% 42% 100% 100% 23% 35% APA 35% 38% 100% 100% 35% 38% Source: Company data, Credit Suisse estimates The wider crude differential adversely impacts cash flows at a minimum and potentially constrains Permian oil production growth. While oil volumes in theory could be moved out of basin on truck and rail, both of these takeaway options are limited by logistics, particularly with already high trucking demand and road congestion. Combine the logistical bottleneck with the negative cash flow impact from wider basis offsetting the benefit of higher WTI/Brent oil prices, basin-wide Permian oil growth could be more tempered than our current estimates. Constraints could also be caused by gas pipeline takeaway starting in 2H18 putting additional pressure on basin production levels. That said, we believe operators (at least names in our coverage universe) are unlikely to change their 2018 drilling programs given pre-planned development schedule and service agreements. Many of the public E&Ps also have wellhead flow assurance with volumes on oil gathering systems and some with long-haul marketing contracts. As such, we expect wider Midland differentials to only result in a financial impact rather than physical impact in the near term. However, if basis continues to worsen to levels near our bear case scenario, we wouldn t be surprised if select E&Ps temper acceleration plans in 2019 if it becomes increasingly difficult or costly to move incremental barrels. Constrained Permian oil growth is also a positive for oil macro fundamentals that should benefit oil-weighted E&Ps with little to no Permian exposure. The Bakken looks particularly attractive given oil differentials have actually improved since start-up of DAPL pipeline last year and the basin is also seeing a technology-driven well productivity uplift. We prefer Outperform-rated CLR, MRO, and OAS for Bakken exposure. The DJ should also be benefiting from new processing capacity addition starting this summer, and we prefer XOG for DJ exposure. Since early Feb, Permian-levered E&Ps have lost 2-3% of cash flow on average due purely to widening of Midland basis differential. Compared to Midland- WTI differential of ~$1/Bbl in early Feb when most E&Ps provided 2018 guidance, cash flow for the Permian E&Ps have already came down by 2% and 3%, respectively, on average even with the basis hedge protection. Our universe of Permian E&Ps has 42% of 2018 Permian oil volumes protected by basis hedges dropping to 13% in At our bear case differential scenario, there s another 1-3% downside to our cash flow estimates. E&P & Refiners 9

10 E&P & Refiners 10 Figure 8: Oil marketing strategy by company and pipeline transportation agreements Part I Primary Market Oil Value Chain Oil Takeaway Details Pricing Ticker Exposure Oil Gatherer Oil Marketer FT Agreements FT Commentary APC Gulf Coast Anadarko Petroleum APA Midland APA gathers Alpine High, uses third parties in Midland and Central Basin Platform CDEV Midland Essentially all piped on Oryx CPE Midland 90+% on pipe using primarily Medallion, also Plains and Enterprise CXO Midland Various providers (noted Plains, Oryx, etc.) DVN >50% Gulf Coast Multiple parties including Targa, DCP and Enterprise EGN Midland ~90% on pipe, primarily Plains EOG FANG >30% Gulf Coast >20% Cushing ~33% Midland Midland Source: Company data, Credit Suisse estimates EOG Resources 90+% on pipe, company-owned gathering system in Spanish Trail and on the Delaware asset Both APC and third parties market oil volumes depending on if they're selling at end market or at in-basin regional facilities Sells at wellhead to marketers or midstream companies that hold FT out of the basin Sells at the wellhead to marketers, primarily BP, Shell, OXY For Medallion, has agreements with large purchasers who buy at the wellhead and moves crude to various markets on CPE's Medallion capacity Has FT to Houston refineries and Gulf Coast for exports on 50% of Delaware production; won't disclose which pipes None in place for Permian NA NA Has agreements in place such that by YE19, they will have 100% covered by FT Have sales contracts with third parties that own FT to get flow assurance to move oil out of the basin but APA will realized Midland prices Have 6mo-1yr sale agreements with marketers to get to Midland. Gets Midland pricing less $2.25/Bbl gathering fee on Oryx Have 40 MBbld of term gathering contract to primary delivery points in Midland (Crane, Midland, Colorado City, local refineries) CXO sells at wellhead; various midstream companies market the crude No long-haul pipeline FT Evaluating future long-haul pipelines on a case by case basis DVN as well as multiple parties including Targa, DCP and Enterprise Sells at the wellhead to marketers, primarily Plains Markets directly to refineries Sells at the central delivery point to third parties including Shell, etc. Half of production in has FT on Longhorn to Houston NA Has FT for 20% of production to Gulf Coast, >30% to Cushing NA Looking to add additional FT but nothing to announce yet NA Prefers to sign 3-5 year agreements believes midstream operators will ultimately overbuild Have contracts in place to get to Midland 13 April 2018

11 E&P & Refiners 11 Figure 9: Oil marketing strategy by company and pipeline transportation agreements Part II Primary Market Oil Value Chain Oil Takeaway Details Pricing Ticker Exposure Oil Gatherer Oil Marketer FT Agreements FT Commentary JAG Midland Majority on pipe; sell at the wellhead NA NA NA NBL LPI currently Midland Gulf Coast post % Gulf Coast OXY Gulf Coast OXY PDCE PE PXD QEP Midland ~33% Gulf Coast ~33% Cushing ~33% Midland Gulf Coast ~75% Cushing or Gulf Coast ~25% Midland NBLX gathers and delivers to Central Gathering Systems (CGS); also uses Oryx ~81% on pipe at YE17; company-owned gathering system Most on truck currently; will have company owned gathering system at Block 4 (Eastern Del.) in 2H18 Co-anchor on Medallion gathering system; 90% is on pipe, likely moving up to ~95% before YE18; short haul trucking cost is ~$2/Bbl Occidental, Sunoco, and Plains purchase crude at tank batteries and transport to centralized locations All on pipe in Mustang Spring, where bulk of current production. Uses company-owned local gathering to central delivery points SM Midland 85-90% on pipe, primarily third party WPX XEC Primarily Gulf Coast Midland Source: Company data, Credit Suisse estimates Mostly on pipe via 50% owned 125 MBbld crude oil gathering system in Stateline area; volumes outside of the region are trucked Noble Energy and Plains All American Sells at the wellhead to marketers - Shell and BridgeTex Occidental Energy Marketing, Inc. (OEMI) Sells at the wellhead to various third party marketers Shell and other undisclosed third parties Pioneer Natural Resources Sells at the wellhead to various 3rd party marketers Sells at the wellhead to various third party marketers WPX Energy Culberson/White City: Plains Sells at the wellhead to Plains, Energy Reeves: Energy Transfer Transfer, and other third party Overall: ~30% of 2018 volumes being trucked marketers FT from CGS to Crane, TX on Alliance Pipeline (owned 50/50 with NBLX and PAA) NA 95 MBbld of firm sales for with multiple marketers, not directly on pipelines; some are long term agreements and ~44 MBbld is take-or-pay agreements, providing flexibility More than 160 MBbld (>90% of production) on Magellan, BridgeTex, Cactus, and Permian Express to the Gulf Coast; through , they have 70-80% of production on FT NA NA NA Has an acreage commitment for up to 75 MBbld on Epic Pipeline (440 MBbld pipe to Gulf Coast) when it comes on line in late 2019; at that point, it will being selling its oil to the Gulf Coast rather than Crane, TX 30 MBbld on Medallion; sells 19 MBbld to Shell (agreement through Jun'20) where they can elect Gulf Coast pricing, 10 MBbld on BridgeTex which provides a basis type hedge between Midland and Gulf Coast (through 1Q25); and 10 MBbld of Mid-Cush basis swaps at ~$0.56/Bbl in 2018 Has 600 MBbld of FT; includes 200 MBbld on Cactus (Gulf Cost), 200 MBbld on BridgeTex (Houston), 50 MBbld on Longhorn (Houston), and 20 MBbld on Sunoco, with future FT on Cactus II and Grey Oak; most agreements extend through 2025 NA Pricing exposure is evenly split among Gulf coast, Cushing, and Midland benchmarks; has ~11.4 MBbld of Mid-Cush hedges in 2018 to protect volumes that are exposed to Midland Realize Gulf Coast prices less $3/Bbl cost of FT so realizations improving with MEH- Midland diff of $7.50/Bbl currently. Have contracts in place to get to Midland, and have pricing agreements with 3rd party offtakers on ~75% of volumes off of out of basin benchmarks NA Current : ~30 MBbld to Corpus Christi, 5 MBbld to Cushing from Midland. Post Oryx/Nov 18: Will have ~125 MBbld of capacity to Midland (up to 100 MBbld on Oryx), of which MBbld can move from Midland to Corpus Christi Long standing, and renewable, marketing arrangement with Plains who takes XEC oil volumes (up to 60 degree API) with no deduct 13 April 2018

12 Implication for Refiners Within refiners DK, HFC and ANDV will be the beneficiaries of blowout in Permian diffs. VLO and PSX also benefit from the blowout but given Permian makes up ~5% of total crude slate the benefits are limited. Historically the best way to play widening diffs was WNR with ~100% Permian refining capacity (prior to WNR acquiring NTI). With the acquisition of WNR by ANDV, DK offers the most pure-play upside from blowout in diffs. With ~207 kb/d of directly sourced Permian crude (69% of total capacity), DK is most leverage to midland diffs blowout. We raise DK PT to $54 (+$7/shr), HFC s PT to $57 (+$4/shr), and ANDV PT to $142 (+$2/shr) to account for higher Permian diffs. Figure 10: Leverage to Permian Diffs kb/d DK ANDV HFC VLO PSX 80.0% 70.0% 60.0% 50.0% 40.0% 30.0% 20.0% 10.0% 0.0% % of Throughput Permian Crude (kb/d) % of Throughput Source: Company data, Credit Suisse estimates We would like to highlight that Permian basin refineries are best configured to run lighter crudes. While Gulf Coast refining capacity (8,300 MBbld) dwarfs Permian refining capacity (~450 MBbld), majority of these are not configured to run crudes grades of 45 and above API gravity. As shown in fig 11 and 12, a high percentage of the crude oil growth in both Texas and New Mexico has been light or super light oil (greater than 45 API). While a number of pipeline projects have been announced to move crude from Permian Basin to Gulf Coast, these projects are struggling to get firm commitments as Gulf Coast refineries need more medium and heavy sour barrels vs. super light crude. Figure 11: New Mexico Crude by API Kb/d Jan-2015 Mar-2015 May-2015 Jul-2015 Sep-2015 Nov-2015 Jan-2016 Mar-2016 May-2016 Jul-2016 Sep-2016 Nov-2016 Jan-2017 Mar-2017 May-2017 Jul-2017 Sep-2017 Nov-2017 Jan-2018 < 30 API API API > 50 API Figure 12: Texas Crude by API Kb/d Jan-2015 Mar-2015 May-2015 Jul-2015 Sep-2015 Nov-2015 Jan-2016 Mar-2016 May-2016 Jul-2016 Sep-2016 Nov-2016 Jan-2017 Mar-2017 May-2017 < 30 API API API > 50 API Jul-2017 Sep-2017 Nov-2017 Jan-2018 Source: DOE, Company data, Credit Suisse estimates Source: DOE, Company data, Credit Suisse estimates E&P & Refiners 12

13 Below we show the API gravity of oil input to the Gulf Coast refining system (both Texas and Louisiana). Texas GC is currently running crude with weighted average API gravity of Degrees (heavier or denser crudes). Louisiana GC is currently running crude with weighted average API gravity of Degrees. Gulf Coast refining system has already made adjustment to lighten feed slate and is absorbing incremental shale barrels. However, as most of the Permian volume growth is in light and super light barrels, Gulf Coast refiners near-term ability to absorb incremental light crude will be somewhat limited. That s one of reasons new pipeline projects are putting an upper limit of API specifications (45 degree) and would not carry super light crude. Figure 13: TX Gulf Coast API of Crude Oil Input Figure 14: LA Gulf Coast API of Crude Oil Input API Jan-2000 Jan-2001 Jan-2002 Jan-2003 Jan-2004 Jan-2005 Jan-2006 Jan-2007 Jan-2008 Jan-2009 Jan-2010 Jan-2011 Jan-2012 Jan-2013 Jan-2014 Jan-2015 Jan-2016 Jan-2017 Jan-2000 Jan-2001 Jan-2002 Jan-2003 Jan-2004 Jan-2005 Jan-2006 Jan-2007 Jan-2008 Jan-2009 Jan-2010 Jan-2011 Jan-2012 Jan-2013 Jan-2014 Jan-2015 Jan-2016 Jan-2017 API Gravity 20 yr Avg. API Gravity 20 yr Avg. Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates We would also like to highlight that while there are some pipeline projects that are providing incremental capacity to Cushing (PAA Sunrise extension), PADD 2 is also running crudes with weighted average API gravity of 33. While PADD 2 refiners can always use incremental barrels to blend with heavier WCS, the demand for straight run incremental lighter barrels is limited. PADD 5 refineries are running a significantly heavier feed slate (API gravity 27). That was one of reason multiple pipeline operators hoping for successful open seasons for pipe capacity from Permian to West Coast never got firm commitments. Figure 15: PADD 2 API of Crude Oil Input Figure 16: PADD 5 API of Crude Oil Input Jan-2000 Jan-2001 Jan-2002 Jan-2003 Jan-2004 Jan-2005 Jan-2006 Jan-2007 Jan-2008 Jan-2009 Jan-2010 Jan-2011 Jan-2012 Jan-2013 Jan-2014 Jan-2015 Jan-2016 Jan-2017 Jan-2000 Jan-2001 Jan-2002 Jan-2003 Jan-2004 Jan-2005 Jan-2006 Jan-2007 Jan-2008 Jan-2009 Jan-2010 Jan-2011 Jan-2012 Jan-2013 Jan-2014 Jan-2015 Jan-2016 Jan-2017 API Gravity 20 yr Avg. API Gravity 20 yr Avg. Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates Given the mismatch between the quality of crude produced (very light) and quality desired (heavier), we believe incremental barrels will have to clear via exports. Post 2020, we see a strong demand growth for these lighter barrels as global refining system makes adjustments for IMO We estimate 5-7% global refining capacity will have to switch to lighter feed slate to cut resid output. E&P & Refiners 13

14 Permian Supply/Demand Considerations Where we stand on production Production Outlook: Based on bottoms-up E&P projections and top-down estimates, we see Permian basin crude production growing from 3.0 MMBbld currently to 3.45 MMBbld by YE18 and over MBbld annually over the next several years to >5 MMBbld by This view is supported by aggregate growth from the 22 companies with Permian exposure in our coverage universe (including CVX and XOM), which can in aggregate add ~560 MBbld of production in 2018 vs (540 MBbld 4Q18 vs 4Q17). Longer term, we estimate the Permian will deliver over 75% of US oil growth over the next 5 years. And with just 17 counties in west Texas driving the lion s share of supply growth to meet global demand, investor concerns regarding infrastructure bottlenecks that could delay this growth have been a nagging concern that have recently been thrust to the forefront as the recent $4-5/Bbl widening in the Midland-Cushing basis has emphasized the arrival of pipeline takeaway bottlenecks. Figure 17: Permian Crude Oil Production 4,800 4,000 Figure 18: Permian Rig and Completion Activity Level 2, Permian Total Oil Production (MBbld) 4,400 4,000 3,600 3,200 2,800 2,400 3,600 3,200 2,800 2,400 2,000 1,600 CS Universe US Oil Production Aggregate Quarterly Well Completions 1,800 1,600 1,400 1,200 1, Average Rig Count 2,000 1, Q17 2Q17 3Q17 4Q17 1Q18 2Q18 3Q18 4Q per 2020 per Quarter Quarter CS Universe Completions Rest of Permian Completions 0 CS Universe Aggregate CS Forecast EIA Total Rigs CS Universe Rigs Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates Near-term milestone / catalyst: What to watch over the next 18 months? We estimate potential for another ~2.6 MMBbld of pipelines to come online by YE20 between greenfield additions and brownfield expansions. However, we believe not all of these projects will be built. We also outline below the key milestone projects/events to watch over the next 18 months that are critical to the Permian supply/demand balance. Green Field Capacity Additions: Looking at just greenfield proposals for the basin, we estimate 1.9 MMBbld of new capacity could come online between YE17 and YE20. Following Midland-Sealy which just recently fully ramped to capacity, we have highest conviction in EPIC and Cactus II (altogether 1.4 MMBbld) plus one more FID from a major pipe through the medium-term. Brownfield Expansions / Conversions: Beyond these larger additions, we are tracking four other major expansions or conversion of existing assets that could add ~560 MBbld of incremental capacity (ETP, EPD, PAA projects). This is a smaller bucket but we see lower capex needs and construction risk increasing their likelihood of success. 1. Borger re-start: Borger refinery ( 50/50 JV between PSX and CVE), was supposed to be fully up and running at this point but hit a couple of snags during its re-start process as it was exiting planned turnaround. It is not uncommon for refineries exiting planned turnarounds to run into minor issues. We expect the refinery will be fully up and E&P & Refiners 14

15 running by April 20, The Borger restart will create incremental +120 MBbld demand for Permian Basin crude. 2. Midland-Sealy Ramp: We believe Enterprise Product Partners Midland Sealy pipeline is already at full capacity of ~500 MBbld. There is little or no spare capacity on this line and it will not change S/D balances in material way on a go forward basis. 3. BridgeTex & Sunrise extension / expansion: Between these two brownfield expansions, an incremental 160 MBbld of takeaway capacity will be added in 1Q19. However, BridgeTex starts at Colorado City, and getting crude from Delaware and Midland basins to Colorado City will itself be a challenge. We believe expanded BridgeTex (440 MBbld) will run below capacity 85-90% utilization given its point of origination (current utilization is below 80%). Sunrise Extension on the other hand terminates into Cushing and not Gulf Coast. Demand for super light crude at Cushing is limited, and we expect this pipeline will also run below capacity. 4. Cactus 2 in service in 3Q19: We believe PAA s 585 MBbld Cactus 2 will be the next large pipeline expansion to materially add takeaway capacity in the Permian. We expect line fill to start in early 3Q19, with the pipe reaching capacity by end of 3Q Permian Express 3 Phase 2 in service in late 3Q19: While we don t have confirmation of a successful open season, we see a high probability this 200 MBbld project does move ahead. We expect line fill to start in late 3Q19, with the pipeline reaching capacity by end of 4Q EPIC in service in 4Q19: By 4Q19, the combination of the Cactus II and the 590 MBbld EPIC Pipeline should materially boost Permian Basin Pipeline takeaway capacity to 5.3 MMBbld. With production forecasted to be 4.2 MMBbld, alleviating all potential bottleneck issues and driving diffs back to transportation economics. 7. Results of open season: We are closely watching the results of open season for BPL s South Texas Gateway (400 MBbld), MMP s Permian to Corpus pipeline (350 MBbld) and PSX/ ENB Grey Oak pipeline (385 MBbld). We believe of these three projects, Grey Oak has the highest probability of a successful open season. We are less optimistic on both BPL s and MMP s projects. Any news of successful open season (for any of the 3) will help alleviate investor s long-term concerns even if they don t immediately change the S/D dynamics. How wide can Mid-Cush differential get? Figure 19: Mid-Cush price deck scenarios and strip Mid-Cush Basis ($/bbl) Our bear case assumes the Midland-Cushing spread blows out to $12/Bbl in 4Q18 and 1Q19. Factors could drive wider spreads: 1) Planned / unplanned refinery downtime in the Permian; 2) Delays that push back the in service date of pipelines; 3) Pipelines refusing to carry super light crude that fails to meet pipeline specs; 4) Permian production exceeding our estimates. We believe any one or combination of these factors could force producers to rail crudes to the East or West coasts which would cost them $12-$14/bbl. 1Q18 2Q18 3Q18 4Q Q19 2Q19 3Q19 4Q Base Case ($0.37) ($5.00) ($6.00) ($8.00) ($4.87) ($8.00) ($6.00) ($2.00) ($1.00) ($4.22) ($0.50) Bear Case ($0.37) ($6.00) ($8.00) ($12.00) ($6.63) ($12.00) ($8.00) ($4.00) ($2.00) ($6.47) ($0.75) Bull Case ($0.37) ($3.00) ($4.00) ($5.00) ($3.11) ($5.00) ($3.00) ($1.00) ($0.50) ($2.36) ($0.50) Mid-Cush Strip ($/bbl) Early Feb'18 ($0.35) ($0.78) ($1.46) ($1.83) ($1.11) ($1.52) ($1.03) ($0.57) ($0.34) ($0.86) ($0.10) Current ($0.37) ($4.35) ($5.53) ($5.04) ($3.84) ($4.58) ($4.15) ($2.39) ($1.24) ($3.08) ($0.65) Source: Credit Suisse estimates E&P & Refiners 15

16 We outline factors that could cause a bear case scenario where differentials reach ~$12/Bbl: 1) Planned / Unplanned refinery downtime in the Permian. Permian refineries consume ~450 MBbld of locally produced crudes. As we have seen with the recent downtime at Borger, even a single refinery downtime can cause crude to back up in the Permian basin driving wider spreads. At this point we are only tracking a 34 day planned turnaround EL Paso refinery in late However, any unplanned downtime at 1) El Paso, 2) Gallup, 3) Borger, 4) Big Springs and 5) Navajo could drive a spike in midland-cushing spreads. We would like to highlight that amongst these, Borger is the highest risk refinery which has history on disproportionality high unplanned downtime. While PSX management has taken a number of measures to improve reliability of this refinery, the results have been mixed. 2) Logistics logjam. Trucks, trucks, trucks.... it takes a lot of trucks to supply the sand to frack a well. Assuming an average Permian well (7,500 foot horizontal lateral), it will require 15 million pounds of sand (1,900 pounds per foot) to frack a well over the course of seven days. That works out to 41 truck turns per day for sand for a single well. And with our forecast of 5,300 well completions this year, the industry will average 4,170 truck turns per day.... equal to ~40 miles of truck stood end-to-end; this compares to our 2017 estimate (when industry was employing shorter laterals and less well completion intensity) of 3,255 truck turns per day. No wonder truckers have seen a nearly 30% increase in wages to six figures! Obviously, the material increase in trucks is causing traffic jams, tearing up roads, challenging logistics, and increasing the risk of companies missing well completion and production targets. 3) Delays that push back the in service date of pipelines. We believe delays could happen due to following 3 reasons: 1) Longer time to get firm commitment and conclude open seasons; 2) Permitting delays; and 3) Construction delays. While Plains Cactus 2 has secured firm commitments, a number of these pipes are still in Open Season. With the number of options available, both producers and refiners are taking their time to decide what s the optimal solution for them. We expect delays in successful completion of open season that will cause startup delays. 4) Pipelines refusing to carry super light crude that fails to meet pipeline specs. As Permian crude slate continues to get lighter, it becomes virtually unusable for a number of Gulf Coast medium to heavy complexity refineries. Even Permian refineries configured to run light crudes have historically rejected crudes with API Gravity over 46. As conventional production is replaced with unconventional, the volume of higher gravity crude available for blending that allowed super light Permian crude to blended up and move via pipelines continues to fall. Refiners that will commit to buying crudes coming to Gulf Coast will put a upper limit on the API density of crude that will move through pipelines. As such, there is a possibility that Permian production remains trapped within the Basin despite available pipeline capacity driving a blow in diffs. 5) Permian production exceeds expectation despite takeaway bottleneck. Our Base case assumes Permian production reaches ~3.45 MMBbld by YE18 and over ~650 MBbld annually over the next several years to >5 MMBbld by However, if operators are unfettered by the recent midland basis blowout, rig addition could exceed our estimates (we estimate rig count staying flattish for the year from current level of 444 rigs). Well productivity could also come in above our expectations. Given the already precarious balance today, any incremental supply would put more pressure on Permian differentials. E&P & Refiners 16

17 E&P & Refiners 17 Figure 20: Permian Basin Crude Takeaway Supply/Demand Balance Takeaway Capacity (kb/d) Q18E 2Q18E 3Q18E 4Q18E 2018E 1Q19E 2Q19E 3Q19E 4Q19E 2019E 1Q20E 2Q20E 3Q20E 4Q20E 2020E 2021E 2022E Pipeline Operator Origination Destination Amdel Pipeline ETP Midland, TX Nederland, TX Basin Pipeline + Expansion PAA Colorado City, TX Cushing, OK Centurion Pipeline OXY Permian Cushing, OK Line 80 PSXP Hobbs, NM Borger, TX WA Line PSXP Goldsmith, TX Borger, TX Wink Pipeline KMI Scurry, Texas El Paso, TX West Texas Gulf ETP Colorado City, TX Longview, TX West Texas Gulf Exp - Houston Access ETP Colorado City, TX Houston, TX West Texas Gulf Exp - Longview Access ETP Colorado City, TX Longview, TX West Texas Gulf Exp - Nederland Access ETP Colorado City, TX Nederland, TX Longhorn Reversal + Expansion MMP El Paso, TX Houston, TX BridgeTex Pipeline + Expansion MMP Colorado City, TX Houston, TX Cactus Pipeline PAA McCamey, TX Gardendale, TX Permian Express I ETP Midland, TX Nederland, TX Permian Express II ETP Midland, TX Nederland, TX Permian Express III Phase 1 ETP Midland, TX Nederland, TX Permian Longview & Louisiana Extension ETP Midland, TX Longview, TX Midland to Sealy Pipeline EPD Midland, TX Sealy, TX Lone Star 12-Inch W. TX Pipe Conversion (Deferred) ETP Midland, TX Corsicana, TX Freedom Pipeline (Cancelled) KMI Permian Los Angeles, CA Total Pipeline 2,407 2,842 2,992 2,992 2,992 2,992 2,992 3,032 3,032 3,032 3,032 3,032 3,032 3,032 3,032 3,032 3,032 3,032 3,032 Potential Expansions Midland to Sealy Pipeline Expansion EPD Midland, TX Sealy, TX Permian Express III Phase 2 ETP Midland, TX Corsicana, TX TexStar Midstream Logistics EPIC Pipeline EPIC Orla, Pecos, Crane, & Midland, TX Corpus Christi, TX Cactus II PAA Permian Basin Corpus Christi, TX Sunrise Extension + Loop (expandable to 200k bpd) PAA Midland / Colorado City, TX Wichita Falls, TX Grey Oak Pipeline (In Service: 2H 2019) PSX / ENB West Texas Corpus Christi, TX Permian to Corpus Christi MMP Permian Basin Corpus Christi, TX NGL Pipeline Conversion EPD Permian Basin Mont Belvieu, TX South Texas Gateway (expandable to 600k bpd) BPL Permian Basin Corpus Christi, TX Permian-to-Nederland (mentioned on 4Q17 call - no credit) ETP Permian Basin Nederland, TX Total Potential Expansion Opps ,105 1,695 1,695 2,080 2,280 2,630 2,630 2,630 2,630 2,630 Refineries Operator Location % of Capacity El Paso (via KMI Wink Pipeline) ANDV El Paso, TX 100% Gallup ANDV Gallup, NM 100% Borger (Net from Permian) PSX Borger, TX 100% Big Spring Refinery DK Big Spring, TX 100% Navajo Refinery HFC Navajo, NM 100% Total Refineries TOTAL TAKEAWAY 2,855 3,290 3,440 3,640 3,640 3,640 3,640 3,800 3,800 4,585 5,175 5,175 5,560 5,760 6,110 6,110 6,110 6,110 6,110 Production and Inflows (kb/d) Q18E 2Q18E 3Q18E 4Q18E 2018E 1Q19E 2Q19E 3Q19E 4Q19E 2019E 1Q20E 2Q20E 3Q20E 4Q20E 2020E 2021E 2022E Production Permian Crude Production 2,022 2,444 3,006 3,082 3,246 3,405 3,169 3,606 3,811 3,991 4,182 3,897 4,357 4,515 4,668 4,821 4,590 5,224 5,864 Annual Increase (+- kb/d) Total Production 2,022 2,444 3,006 3,082 3,246 3,405 3,169 3,606 3,811 3,991 4,182 3,897 4,357 4,515 4,668 4,821 4,590 5,224 5,864 Pipeline Operator Origination Destination Bobcat Pipeline ANDV Mason Station, TX Wink Station, TX Total Inbound Pipelines TOTAL SUPPLY 2,147 2,569 3,131 3,207 3,371 3,530 3,294 3,731 3,936 4,116 4,307 4,022 4,482 4,640 4,793 4,946 4,715 5,349 5,989 Source: Company data, Credit Suisse estimates 13 April 2018

18 Greenfield Projects With the backdrop of ~3.1 mmb/d of pipeline capacity already in-service, we are following ~2.9 mmb/d of new potential Greenfield projects. EPD Midland-to-Sealy: 500 kb/d by early 2Q18 to Houston (now at full capacity) EPIC Crude Pipeline: 590 kb/d in 2019 to Corpus PAA Cactus II: 585 kb/d in 3Q19 to Corpus ETP Permian-to-Nederland: 400 kb/d (assumed) to Nederland BPL South Texas Gateway: 400 kb/d Corpus and Houston (low probability) PSX / ENB Grey Oak Pipeline: 385 kb/d in late 2019 to Corpus MMP Permian: 350 kb/d to Corpus / Houston (low probability) Figure 21: Permian Greenfield Pipeline Map Source: Company data, Credit Suisse estimates, SNL Figure 22: Disclosed Greenfield Permian Pipeline Projects Incremental Included in Project Company Destination Capacity (kb/d) Startup CS Model CS View Midland Sealy EPD Houston 200 In Service Y Line running at full capacity of 500kb/d) Cactus II PAA Corpus 585 3Q 2019 Y Fully Committed EPIC Private Equity Corpus 590 4Q 2019 Y Committed Gray Oak PSX & ENB Corpus & Houston 385 1Q 2020 Y Still in Open season (some risk of firm commitment) Permian to Nederland ENB Houston / LA 400 NA N No Clarity on timing (excluded in our model) Permian to Corpus MMP Corpus & Houston 350 2Q 2020 Y Extended Open season ( low success probability), still included in our model South Texas Gateway BPL Corpus & Houston 400 NA N Extended Open season ( low success probability), excluded in our model Total (kb/d) Source: Company data, Credit Suisse estimates E&P & Refiners 18

19 Midland-to-Sealy Pipeline (EPD) The Midland-to-Sealy Pipeline is operated by Enterprise Product Partners and now has a capacity of 500 kb/d after an incremental 200 kb/d in takeaway got added. Midland-to- Sealy can move WTS, WTI, light WTI and condensate and is intended to target end markets on the Gulf Coast. Current Status: Based on our conversations with the company, Midland-to-Sealy is now fully ramp to 500 kb/d. Shippers: EPD ships a significant amount of their own volumes from their West Texas Gathering system. EPD also has committed capacity from CVX and ECA. Infrastructure to Leverage: Midland-to-Sealy is not dependent on 3rd party volumes as it feeds from EPD s West Texas Gathering system and is connected to EPD s Midland crude terminal. It is connected to EPD s 6.4 mmbbl ECHO oil terminal on the Gulf Coast via an existing 1 mmb/d Rancho II connection. The project can also leverage EPD s export marine terminals in Beaumont and in the Houston Ship Channel. Our Thoughts on the Project: Already online, no risk to the project but no spare capacity. Figure 23: Project Map Source: EPD May 2017 Investor Presentation E&P & Refiners 19

20 EPIC Pipeline (ARES Capital) The 590 kb/d EPIC Pipeline is capitalized by the private equity firm ARES Capital. EPIC was formed in 2017 by the management team behind TexStar Midstream. The proposed pipeline would run from Orla, TX in the Delaware and Midland, TX in the Midland Basin to the Gardendale, TX hub in the Eagle Ford, and then subsequently run to Corpus Christi. Current Status: On December 21, 2017 EPIC announced that they had received sufficient shipper interest via the 10-day open season to move forward with the project and begin construction. The planned in-service date is Shippers: 75 kb/d of capacity has been reserved by NBL via an acreage dedication for substantially all of its gross acre position in Reeves Co in the Delaware. This relationship includes a strategic partnership in which NBL or NBLX has the option until January 2019 to contribute equity for up to a 30% stake in the EPIC pipeline. Another 75 kb/d has been reserved by an additional undisclosed shipper. Infrastructure to Leverage: The project will include terminals in Orla, Pecos, Crane, Wink, Midland, Helena, and Gardendale, TX. The pipeline is aimed at Corpus Christi export markets and will run alongside EPIC s NGL Pipeline. Our Thoughts on the Project: We expect 4Q19 startup. Figure 24: Project Map Source: EPIC Pipeline Company Website E&P & Refiners 20

21 Cactus II (PAA) The planned Cactus II system will involve two new greenfield pipes in combination with existing pipelines with total capacity additions of 585 kb/d in crude takeaway (expandable to 670 kb/d). Cactus II will extend from Wink South to McCamey, TX and from McCamey to Corpus Christi. Pipeline origin points include Orla, Wink South, Midland, Crane, and McCamey, TX. Current Status: The initial open season on the pipeline closed in late January with PAA having received enough binding interest to begin construction. The second open season also concluded with the project now fully committed. The project is currently undergoing permitting, right-of-way, and procurement activities with a target ISD of 3Q19. Shippers: PAA has announced that the pipeline is now fully committed with long-term 3rd party contracts amounting to 525 kb/d (425 kb/d in MVCs kb/d in acreage dedications). Another 60 kb/d is reserved for walk-up shippers. Commodities trader Trafigura is also expected to move 300 kb/d on Cactus II (~52% of total planned capacity). Infrastructure to Leverage: Cactus II will expand PAA s current Permian supply-push system and be able to leverage their current crude storage capacity (two terminals located in the Permian and an additional terminal located in Corpus Christi). Our Thoughts on the Project: We expect 3Q19 startup. Figure 25: Project Map Source: PAA 4Q17 Earnings Presentation E&P & Refiners 21

22 Grey Oak Pipeline (PSX / ENB) The Grey Oak Pipeline, jointly operated by ENB and PSX, is planning to provide 385 kb/d from West Texas to the Corpus Christi, Freeport, and Houston, TX markets. Origin points are anticipated to be located in Reeves, Loving, Winkler, and Crane counties in Texas. Current Status: Open season commenced on December 11, The project is expected to come online in 2H19 (although we estimate 1Q20 in our model) and ENB/PSX will evaluate further capacity expansions additions depending on shipper interest levels. Shippers: TBA Infrastructure to Leverage: If completed, the project would leverage PSX s refining capacity. The Sweeny facility in Houston has 247 kb/d of refining capacity and PSX has a total of 743 kb/d levered to the overall Gulf Coast region. Our Thoughts on the Project: While the pipeline is still in open season we are confident it gets enough shipper commitments to proceeds with the project. We expect 1Q20 startup. PSX s moved NGLs for Permian producers through Sand Hills and Southern Hills and also is a major player in gas processing through DCP. These relationships will help PSX secure firm commitments ahead of competitor projects. Figure 26: Project Map Source: ENB December 2017 Investor Update Presentation E&P & Refiners 22

23 Permian to Corpus / Houston (MMP) The Permian to Corpus / Houston pipe intends to move volumes from the Permian + Eagle Ford with origin points in Crane, Orla, Wink, Midland, Helena, and Gardendale, TX. Delivery points will include Crane (to Longhorn), Corpus Christi (MMP terminal), Ingleside, and MMP s Houston Distribution System. Current Status: MMP launched open season for the proposed 350 kb/d (expandable to 600 kbp/d based on shipper demand) pipeline on Dec. 1, 2017 which was set to close Feb. 1, 2018 but was subsequently extended to March 1. Contracts & Tariff Details: While details have yet to be announced, committed shippers will have the ability to move volumes to Corpus or Houston and could designate 3rd parties to use their capacity. An incentive tariff rate will allow committed shippers to aggregate vols to reach higher threshold tiers. Shippers: While specific details have yet to be announced at this time, MMP has indicated that significant interest has been expressed from potential shippers, and the [open season] extension provides these potential shippers additional time to finalize their commitments as of a 1/31/18 press release. Other Infrastructure to Leverage: MMP plans to leverage existing crude storage terminals in the Corpus / Houston region and is also considering other pipeline extensions to further expand their Permian + Eagle Ford footprints. They highlighted the project as providing deep water dock access (MMP Corpus + Seabrook marine terminals and Ingleside terminals). Our Thoughts on the Project: We see a low probability for success of this project. The delays are indicating lack of firm commitments. While we are including this project from our pipeline expansion analysis (3Q20 startup) there is risk that this pipeline does not see the light of the day. Figure 27: Project Map Source: MMP Jan 2018 Investor Presentation E&P & Refiners 23

24 Permian-to-Nederland (ETP) On their 4Q17 call ETP indicated that they were aggressively pursuing a larger project but did not give much in the way of specific details. This proposed Permian-to-Nederland crude line would carry volumes from the Permian Basin to Nederland, TX. Current Status: While the project is still very much in the early planning stages, ETP on the call indicated that they would be looking to bring strategic partners in on this project as well so we believe they are likely actively looking for a JV partner to de-risk the project before announcing further details at this point. Contracts & Tariff Details: TBA Shippers: On the 4Q17 call ETP highlighted that their customers have asked us to do this, to look at a much bigger project and that they have a lot of momentum with a handful of producers, and that is probably the cheapest type of transportation that we can provide. They noted that they are going to build whatever capacity is warranted after those discussions which seems to indicate that the project already has significant producer backing even at this early stage. Other Infrastructure to Leverage: ETP would leverage their 26 mmbbl Nederland, Texas terminal facility which further allows ETP to benefit from an integration standpoint. On the 4Q17 call ETP noted that the project would provide shipper capacity to our storage facilities and pipeline header systems [in Nederland]. Our Thoughts on the Project: Given the lack of details (still in planning and not open seasons) we are not including this project in our analysis. We do not see this pipeline coming online until Figure 28: Project Map Source: Company data, Credit Suisse estimates, SNL E&P & Refiners 24

25 South Texas Gateway (BPL) The South Texas Gateway is a planned 400 kb/d (expandable to 600 kb/d based on shipper interest) pipe that would move crude from the Permian and Gardendale, TX to the Corpus Christi / Ingleside / Houston region along the Gulf Coast. It would begin in Wink and Midland, TX and end at BPL s refining and export facilities in Corpus Christi. Current Status: BPL launched the initial binding open season for the project in late December The project is expected to begin construction in late 2018 with a target ISD early Shippers: TBA Infrastructure to Leverage: We believe South Texas Gateway could work well with BPL s current Corpus asset footprint + buildout. Current Corpus buildout includes expanded storage facilities (specifically for South Texas Gateway / BTP terminal customers), a fifth deepwater dock, and the Buckeye Texas Market Center distribution system. BPL is also evaluating a Permian-to-Corpus NGL long-haul pipeline. Our Thoughts on the Project: We see a low probability for success of this project. We believe producers will prefer to go with midstream producers that have stronger gathering capability. As a result we are excluding this project from our pipeline expansion analysis. Figure 29: Project Map Source: BPL South Texas Gateway Pipeline Company Website E&P & Refiners 25

26 Large Brownfield Expansions / Conversions Beyond the major greenfield proposals, we see three high-likelihood expansions or asset conversions that can add another ~560 kb/d of basin capacity. This is a relatively lower number, but given the low capex requirements needed and lower construction risk, we see this group as offering higher conviction. ETP Permian Express III Phase 2: +200 kb/d by YE19 (est) EPD NGL Pipeline Conversion: +200 kb/d by 1H20 PAA Sunrise Pipeline Expansion: +120 kb/d in 1H19 BridgeTex : +40kb/d 1Q19 Figure 30: Brownfield Expansions / Conversions Map Source Company data, Credit Suisse estimates, SNL Figure 31: Potential Brownfield Expansions / Conversions Project Company Destination Startup Incremental capacity Permian Express 3 Phase 2 Energy Transfers Hosuton / LA 3Q NGL conversion to Crude Enterprise Products Houston 2Q Sunrise Extension Plains Cushing 1Q Bridge Tex Mgellan / Plains Houston 1Q Total 560 Source: Company data, Credit Suisse estimates E&P & Refiners 26

27 Permian Express III Phase 2 (ETP) ETP s Permian Express III Phase 2 is slated to add an incremental 200 kb/d in crude takeaway from their Permian Express Terminal at Garden City, TX in West Texas to their 26 mmbbl Nederland, Texas terminal facility. The project passes through Colorado City and Corsicana, TX before reaching the Texas Gulf Coast. This project is an expansion to the current Permian Express II pipeline as opposed to an entirely new asset. Current Status: Currently under development as per the latest investor presentation but has yet to launch an open season. The 100 kb/d Phase 1 saw a successful open season in 2017 with a targeted in-service of 4Q17 which was met on time. ETP expects to launch an open season for phase 2 soon. Contracts & Tariff Details: Currently on the Permian Express II, the tariff rate is estimated to be ~$2.6/bbl for committed shippers with a walk-up rate of $4.05/bbl so we would expect a similar rate for this project Shippers: The current Permian Express II pipeline has committed capacity from Citgo, XOM, SXL Marketing, and others. Permian Express Partners, the entity which holds the asset, is a joint venture with XOM in which ETP owns 88%. As a JV partner, it would be expected that XOM would continue to support future expansions via committed volumes (including the Permian Express III Phase 2). Infrastructure to Leverage: The Permian Express system is interconnected with other portions of ETP s Permian crude network including SXL Crude, West Texas Gulf, Permian LV & LA Extension, and the Delaware Basin Extension. The Permian Express feeds into their 26 mmbbl Nederland, Texas terminal facility which allows ETP to benefit from an integration standpoint. Our Thoughts on the Project: While the pipeline is still in not in open season we are reasonably confident it gets enough shipper commitments to proceeds with the project. We expect 3Q19 startup. Figure 32: Project Map Source: ETP January 2018 Investor Presentation E&P & Refiners 27

28 Sunrise Pipeline Extension + Loop (PAA) In April of 2017 PAA had announced an open season for an expansion of up to 350 kb/d in incremental crude takeaway on their current Basin system via the Sunrise Extension + Loop. The project originally was intended to originate in Midland / Colorado City, TX and terminate in Cushing, OK. But after an extension followed by a complete reboot of the open season, the project in its current form now is set to begin at PAA s Conan Station in Loving, TX in the Delaware and connect to Colorado City, TX in the Midland before finally terminating at Cushing, OK. The project involves expansion of existing pipeline in addition to greenfield pipe and will add 120 kb/d in takeaway (expandable to 200 kb/d). The Sunrise Expansion on Basin is expected to be in-service 1Q19. Current Status: The original open season that began in April of last year was originally slated to end in May 2017 but was then extended to early June. By mid-june PAA announced a new open season which then successfully concluded by mid-july The Sunrise Expansion is now awaiting permits + regulatory approvals before beginning construction. Other Infrastructure to Leverage: The expansion would deliver into PAA + third party terminals. The Sunrise project will expand PAA s current Permian supply-push system and be able to leverage their current crude storage capacity (two terminals located in the Permian and an additional terminal at Cushing, OK). Our Thoughts on the Project: We believe while the project does provide 120 kb/d of incremental takeaway capacity, since the line terminates at Cushing vs. Gulf Coast, we expect it will be running below capacity. Also the line just removes bottleneck from one region to another. Figure 33: Project Map Source: PAA December 2017 Investor Presentation E&P & Refiners 28

29 NGL Pipeline Conversion (EPD) EPD is planning to convert one of its NGL pipelines into a crude oil pipeline. As of today, they estimate this pipeline will move between kb/d. EPD has three existing NGL pipelines that run from the Permian Basin to the Texas Gulf Coast: the Seminole Blue, Seminole Red, and Chaparral pipelines. The Seminole system has a combined 300 kb/d in NGL takeaway capacity and primarily moves NGLs from the Rockies. Chaparral has 135 kb/d in NGL capacity and is sourced by ~30 natural gas processing plants in the Permian region. EPD also plans to bring the currently under-construction Shin Oak NGL pipeline by 2Q19. All systems move NGLs to Mont Belvieu. Current Status: As per the company, the planned repurposed NGL line is currently supported by strong customer demand. EPD has yet to announce which NGL pipeline will be converted to move crude to the Texas Gulf Coast. Shippers: TBA Infrastructure to Leverage: It is likely that the completed project would leverage EPD s export marine terminals in Beaumont and in the Houston Ship Channel. Our Thoughts on the Project: Final takeaway capacity will depend on which line EPD decided to convert, but capacity could be anywhere from MBPD. EPD will need to complete the Shin Oak NGL pipeline before converting an NGL line to crude service (since it will need to do something with the NGLs). Shin Oak is scheduled to come on line in 2Q We expect 2Q 2020 startup for one of these converted pipelines. Figure 34: Project Map Source: EPD February 2018 Investor Presentation E&P & Refiners 29

30 Select Existing Pipelines Of the ~3.3 mmb/d of current capacity, we include a breakdown of the major pipelines already online (comprising 2.5 mmb/d or ~75% of total). Largest players of the overall list are ETP and PAA (both ~ 1 mmb/d), followed by MMP and EPD (both ~500 kb/d). ETP Permian Express I-III: 450 kb/d (150, 200, 100 kb/d) to Nederland. PAA Basin: 450 kb/d to Cushing MMP/PAA BridgeTex: 440 kb/d to Houston ETP West Texas Gulf: 360 kb/d to Houston / Area PAA Cactus I: 345 kb/d to Corpus MMP Longhorn: 275 kb/d to Houston OXY Centurion: 140 kb/d to Cushing ETP Permian Longview + LA Extension: 100 kb/d to Longview Figure 35: Select Existing Pipeline Capacity Out of the Permian - Map Source: Company data, Credit Suisse estimates, SNL Figure 36: Select Existing Pipeline Capacity Out of the Permian Project Company Destination Base Capacity Remaining Total Capacity (kb/d) Expansion (kb/d) (kb/d) Permian Express I - III Energy Transfer Partners (ETP) Houston / LA Basin Plains All American (PAA) Cushing BridgeTex Magellan (MMP) / Plains (PAA) Houston West Texas Gulf Energy Transfer Partners (ETP) Houston / LA Cactus I Plains All American (PAA) Corpus Longhorn Magellan Midstream (MMP) Louisiana Centurion Occidental Petroleum Cushing Longview + LA Ext. Energy Transfer Partners (ETP) Houston / LA Total 2, ,560 Source: Company data, Credit Suisse estimates E&P & Refiners 30

31 Basin (PAA) PAA s Basin provides Delaware + Midland takeaway to Cushing, OK. The major portions of the asset run from Jal, NM to Wink, TX, then from Midland to Colorado City, TX and from Colorado City to Wichita, TX before heading to Cushing, OK. The pipeline currently moves 250 kb/d and is 87% owned by PAA with EPD owning the remaining 13% portion. The asset came into service pre-2010 and can move WTI, WTS, and condensate. Contracts & Tariff Details: The tariff to Midland is $0.76/bbl but we estimate that moving crude from the Delaware to Cushing, OK can be upwards of ~$2/bbl. As per FERC filings, it currently costs shippers $0.65/bbl to move crude from Conan Station in Loving, TX to Midland, TX and it costs another ~$0.77/bbl to ship from Midland, TX to Cushing, OK, implying an all-in tariff of ~$1.42/bbl. Shippers: The Basin system has committed capacity from PAA s marketing business in addition to Occidental Petroleum Corp (OXY), among others. Infrastructure to Leverage: In addition to providing long-haul takeaway, PAA s Basin provides shippers with access to a vast crude gathering network which feeds into the mainline. We highlight that the pipeline generally doesn t run near full capacity given that many producers prefer to ship barrels to the Houston/Corpus region rather than to Cushing, OK. Figure 37: Asset Map Source: PAA March 2017 Presentation E&P & Refiners 31

32 Centurion (OXY) Centurion Pipeline is 2,900 miles of pipeline extending from South East New Mexico across the Permian to Cushing, OK, providing 140 kb/d of transportation capacity. Centurion targets market centers, third-party connecting carriers and refineries. Centurion Pipeline is a wholly-owned subsidiary of Occidental Petroleum Corp (OXY). The pipeline came into service pre We would warn investors to not confuse the primary long-haul portion of Centurion with the more than ~2,000 miles of gathering pipelines included in the system when citing capacity figures (the overall system is ~2,900 miles of pipeline). Contracts & Tariff Details: We estimate that Centurion charges roughly $1.43/bbl to move volumes to Cushing, OK from the Permian. Shippers: OXY moves a significant portion of their own volumes on the system. Centurion also moves oil for marketers, other producers, refiners, and shippers. As with PAA s Basin, we highlight that the asset generally doesn t run at full capacity given that many shippers prefer to move volumes to the Houston/Corpus areas rather than to Cushing, OK. Infrastructure to Leverage: OXY leverages ~2,000 miles of gathering pipelines included in the Centurion system in addition to roughly 70 truck unloading facilities (out of a total of 120 unloading stations located throughout New Mexico, Texas, and Oklahoma). Figure 38: Asset Map Source: OXY Sept 2012 Permian Basin Overview Presentation E&P & Refiners 32

33 Cactus I (PAA) Cactus I is a 310 mile, 20-inch crude oil pipeline capable of transporting 390 kb/d from McCamey, TX to Gardendale, TX. Initially, Cactus I was designed to transport 200 kb/d, however continued demand from shippers has led to three expansions with the most recent expansion in 3Q17 taking capacity to 390 kb/d (up from 330 kb/d). Cactus is capable of transporting sweet and sour crude grades and it originally came online 1Q15 with 250 kb/d before it was expanded to 330 kb/d in We also highlight that the pipe is capable of moving up to five batches of crude or condensate daily but typical batches are only usually three different grades. Shippers: Plains moves their own volumes on Cactus, while OXY has also committed capacity. Most of the crude on Cactus flows to either refineries in the Corpus Christi area or is loaded onto Jones Act vessels on the Gulf Coast for export. Infrastructure to Leverage: Cactus I can serve the Houston-area market via a connection to Enterprise s South Texas Crude Oil Pipeline. Cactus I also feeds into Three Rivers (where a Valero refinery is located) and Corpus Christi markets, delivering crude into the Eagle Ford JV (660 kb/d system 50% owned by PAA) and other third party facilities. Figure 39: Asset Map Source: EIA E&P & Refiners 33

34 Longhorn (MMP) The Longhorn Pipeline is a 450-mile pipeline system capable of transporting 275 kb/d of crude oil from the Permian Basin to various points in Texas. Shipments originate in Crane, Barnhart, or Midland, TX via either trucks or interconnections with crude gathering systems. Longhorn is owned and operated by Magellan Midstream Partners who acquired the system in 2009 in the pre-shale era days. It was reversed in 2013 to flow in its current direction. Pipe quality limits allow it to move WTI only. Contracts & Tariff Details: Average committed tariff currently ~$2.30/bbl, spot tariff closer tariff closer to $4/bbl. At its recent analyst day, MMP confirmed almost all existing customers have not extended their customer contracts under current terms of 2 years. Shippers: Longhorn has committed capacity from Shell and other E&Ps. Infrastructure to Leverage: Crude on the system feeds into terminals in Houston, TX including MMP s East Houston storage and distribution facility. These assets then connect into refineries in the Houston and Texas City region. Figure 40: Asset Map Source: MMP Company Website E&P & Refiners 34

35 BridgeTex (MMP / PAA) BridgeTex is capable of transporting 400 kb/d from Colorado City, TX to Houston, TX. BridgeTex is a 50/50 JV between Plains and Magellan. The pipeline was recently expanded from 300 kb/d to 400 kb/d with construction completing in 3Q17, however, BridgeTex is slated to add an additional 40 kb/d of capacity by early 2019 with a supplemental open season having launched Nov The original 300 kb/d came online in September The asset can move WTI, WTS, and condensate. Shippers: About half a dozen major shippers are estimated to have long-term contracts on BridgeTex, including anchor shipper OXY. The company expects shipments to be 315 kb/d or slightly higher than committed levels, at average rate of $2.35/bbl. Infrastructure to Leverage: The pipe can leverage storage capacity in Colorado City, TX and at the Magellan East Houston Terminal. It is fed by Basin, Mesa, and Sunrise as oil come from Midland, TX (most production occurs west of Colorado City). Figure 41: Asset Map Source: MMP 2016 Analyst Day Presentation E&P & Refiners 35

36 West Texas Gulf (ETP) ETP s West Texas Gulf (WTG) moves crude from Midland and Colorado City, TX to Longview, TX or alternatively to Goodrich, TX (between Houston and Nederland). The asset was formerly owned by Sunoco Logistics (SXL, now owned by ETP) and came online pre Since commencing operations, the asset has seen three major expansions: 1) +40 kb/d via the Houston Access project in 2014, 2) +30 kb/d from the Longview Access project in 2014, and 3) an additional +40 kb/d via the Nederland Access project in This incremental 110 kb/d brought total takeaway to the current 360 kb/d level. Quality limits allow WTG to only move WTI. Shippers: Medallion Gathering & Processing (private midstream company sold to Global Infrastructure Partners early 4Q17) has connections to WTG via their Wolfcamp Connector System in the Midland Basin which came online in Infrastructure to Leverage: To the west of Longview is Delek s 75 kb/d Tyler, TX refinery which sources WTI crude from West Texas Gulf. The system also leverages ETP s Nederland, TX crude oil terminal. Figure 42: Asset Map Source: ETP / SXL Company Website E&P & Refiners 36

37 Permian Express I and II (ETP) The Permian Express I has 150 kb/d of capacity carrying crude from Wichita Falls, TX to the Nederland crude terminal at the Gulf Coast. Originating at Midland, Garden City, and Colorado City, TX, Permian Express II is capable of carrying 200 kb/d of crude to ETP s tank terminals located at Colorado City and Corsicana, TX. Both Permian Express I and Permian Express II are owned by Permian Express Partners (PEP), a joint venture between SXL (Now ETP) and Exxon Mobil, with a 88%/12% split, respectively. Contracts & Tariff Details: Currently on the Permian Express II, the tariff rate is estimated to be ~$2.6/bbl for committed shippers with a walk-up rate of $4.05/bbl Shippers: The current Permian Express II pipeline has committed capacity from Citgo, XOM, SXL Marketing, and others. Permian Express Partners, the entity which holds the asset, is a joint venture with XOM. As a JV partner, it would be expected that XOM would continue to support future expansions via committed capacity. Infrastructure to Leverage: The Permian Express system is interconnected with other portions of ETP s Permian crude network including SXL Crude, West Texas Gulf, the Permian LV & LA Extension system, and the Delaware Basin Extension. The Permian Express also feeds into their 26 mmbbl Nederland, TX terminal. Figure 43: Asset Map Source: ETP April 2017 Investor Presentation E&P & Refiners 37

38 PELA (ETP) The Permian Longview and LA Extension (PELA) includes a section stretching from Garden City / Midland, TX to Colorado City, TX then to Longview, TX which came online 3Q16 with 100 kb/d as a partnership between Exxon, Sunoco Logistics (now ETP), and energy trader Vitol. The Longview, TX to Anchorage / Baton Rouge, LA section was previously known as the North Line Pipeline which was reversed by operator Exxon in The North Line originally moved barrels from the St. James crude hub in LA to Longview, TX. The PELA system is 85% owned by ETP (formerly SXL) and 15% by XOM. Contracts & Tariff Details: Moving barrels from Garden City, TX to Longview, TX is estimated to be ~$3.75/bbl - $4.44/bbl and shipping barrels from Garden City, TX to Baton Rouge, LA is estimated to be ~$4.55/bbl - $7.44/bbl. Shippers: The pipeline has committed capacity from Exxon, Citgo, Delek, and others. Vitol is likely also a shipper given their original assistance with bringing the pipe to FID. Infrastructure to Leverage: At Longview, TX the asset ties into other pipelines which can move barrels to the north or northeast to feed Mid-Con and Midwest refineries. XOM also operates a 503 kb/d refinery in Baton Rouge, LA which now sources Permian crude via PELA since the 2015 reversal. The line also feeds Delek s 80 kb/d El Dorado refinery in LA. Figure 44: Asset Map Source: ETP/SXL November 2014 Investor Presentation E&P & Refiners 38