Description

We believe Shell Midstream (SHLX) is a collection of ratable, stable assets that have been dragged down with the energy sell-off in general and MLP sell-off specifically.  Despite these characteristics and the recent recovery in oil prices / refined product demand, SHLX remains down ~50% from its highs earlier this year.  SHLX sports a 15% dividend yield and 10% unlevered FCF, virtually zero capex, and an exceptionally supportive sponsor, who is currently waiving $20m/quarter of dividends it is entitled to through Q1 2021.  We think the opportunity exists because a) SHLX has wavered recently in its commitment to its dividend, b) will run a coverage ratio in Q2 below 1x, and c) is owned by investors who don’t understand with any granularity what they own. We expect the business to be substantially more resilient long-term than suggested by their Q2 guidance or by the current valuation.  We see roughly 45-75% upside from here while receiving 15% each year in dividends while we wait. While we generally don’t like MLPs, we happen to like SHLX in particular and think it avoids many of the typical MLP pitfalls that investors frown upon. We provide some perspective on this in the first comment and would be curious to hear others’ views on the topic.

For this investment to work you need to generally believe 3 things: 

1) Shell (the parent) will not go bankrupt and will continue to treat minority unitholders well

2) People will drive again, not necessarily at 2019 levels but more than they did in March-May

3) Long-term oil prices above ~$20

Why We Like SHLX in a Nutshell:

In a world with tremendous uncertainty, we believe the three key assumptions we highlight are relatively easy to underwrite. In return, you receive a business with:

1. ~85% of FCF from highly ratable / safe assets with minimal ongoing capital requirements.  These primarily consist of refinery logistics assets with multi-year take-or-pay contracts from Shell, two large refined product pipeline JVs (Colonial and Explorer), and a set of offshore oil pipelines that we believe are roughly on par with refined product pipelines in attractiveness (more on this later). The assets that are of lower quality are a small portion of total FCF and are mostly slowly declining, low capex assets over time that will generate strong cash flows on the way down.
2. An unusually clean cash flow statement for an MLP, with a long history of actual FCF.  The company has also done nearly no organic growth projects, which we actually view favorably, as we believe most large growth projects definitionally exist because barriers to entry are low and that these assets almost always end up overbuilt (e.g. Permian takeaway capacity).
3. An exceptionally supportive sponsor, which in no particular order:
1. Owns 70% of the equity
2. Collapsed IDRs at certain terms in February 2020, and did not modify those terms at all when the stock subsequently fell nearly 60% between announcement and transaction close
3. Frequent IDR / dividend waivers by the sponsor to keep coverage ratio above 1x
4. Resisted the temptation to drop down low quality assets, and in general has dropped high quality assets at very attractive prices into the MLP
5. Has an incentive to keep stock price so high they can issue equity at the MLP level to buy assets from the sponsor and effectively lower the cost of capital of the parent. 
6. Run by a mgmt team independent of the sponsor, which helps limit conflicts of interest
7. MLP debt is issued by Shell and contains no covenants - this lowers cost of capital for SHLX and makes it very difficult for them to get into a liquidity crisis
4. The vast majority of the weakness in near term results are driven by refined product pipeline JVs, which are irreplaceable assets with flattish volumes that can consistently take price each year.  Based on recent trends in fuel consumption as well as FERC and private co financials (Colonial has public debt), we believe you can underwrite a sizeable rebound in distributable cash flow due to refined product demand down only (10%) in recent weeks vs. (30%+) at the time of their Q1 earnings call.  

Why The Opportunity Exists

Shell has several characteristics that we believe are creating its current mispricing:

1. Shell eliminated the IDRs in late February, which opens up a wider portion of MLP investors, many of whom definitionally will not invest in entities with IDRs
2. Shell the parent cut its dividend for the first time since WWII in Q1, and SHLX suggested it would evaluate its dividend on a quarterly basis on its Q1 earnings call, leading people to think they might cut their dividend as well.
3. Poor guided Q2 results that will result in dividend cover below 1x, and limited disclosure as to when this may recover.
4. Collection of unique assets for which there are not many pureplay comps and which require a substantial amount of diligence and work to form a unique view.
5. Majority of assets do not have true MVCs - though we would argue they don’t have MVCs because the assets are so good (e.g. refined product pipe) that in most cases you don’t need them.
6. Historically has traded at a mid-teens EBITDA multiple, hasn’t been cheap enough until recently for “MLP vultures” like us and those on this board who find these situations attractive.
7. Multiple JVs, conservative accounting of EBITDA relative to peers make valuation complicated

Asset Overview

SHLX owns four buckets of assets: refined product pipelines (18% of UFCF), refinery-based assets (20% of UFCF), offshore crude pipelines (56% of UFCF), and a small hodge-podge of other assets (6% of UFCF). We break out the asset financials in detail in the appendix.

We believe 84% of cash flow (refined pipelines + refinery-based assets + ~80% of offshore crude pipelines) is from high quality, long-lived, ratable revenue. Below we detail the asset collection but spend the most time on offshore pipes, which we view as the most mis-understood assets.

Refined product pipelines (18% of UFCF): 

Assets: Colonial, Explorer

Refined product pipelines in the US are virtual monopolies with insurmountable barriers to entry and 2-3% pricing power per year.  They are the most efficient way for refined products to get from refinery locations to demand centers.  For these assets to cash flow, you need to believe that refined product consumption will generally be flattish to down modestly. 

At the height of COVID in late March, we believe refined product volumes fell as much as 50%.  Today, volumes are down closer to 10% YoY.  We think the bulk of the impact Shell will feel will be in Q2, even if covid resurges again.  Because these are JVs who pay dividends as a percent of net income, and because D&A is high, we expect cash distributions to be down sharply in Q2 as net income falls more than EBITDA and quickly rebounds given the high incremental margins and much higher run-rate consumption of refined product (as well as JV level debt).

Shell owns stakes in both Colonial & Explorer.  Explorer provides refined product from the Gulf Coast to the Chicago area and is a solid asset.  Colonial is probably the best refined product pipeline in the US, carrying product from the gulf coast through the Midwest and up to the Northeast.  For years it operated near capacity so much so that line space on Colonial actually trades (suggested the tariff is artificially low).

Refinery-based assets (20% of UFCF):

Assets: Refinery gas pipelines, Norco assets, Terminal assets, ~20% of Zydeco, Bengal Pipeline

These are very safe logistics assets that service one or multiple refineries -- pipelines, oil product storage terminals, and other inside-the-gate refinery logistics assets (e.g. truck racks, docks, etc). 

~85% of cash flow (Refinery gas pipelines, Norco assets, Terminal assets) is virtually guaranteed for the next 10+ years. These assets are underpinned by take-or-pay contracts with >10 years to expiry, where Shell is the counterparty. The remaining 15% of cash flow is from pipelines that service refineries without take-or-pays / MVCs. Bengal pipeline is the 4 major Louisiana refineries to get their refined product outputs to their main markets. The Zydeco “other” pipelines are trunk lines that bring in crude to refineries off the Zydeco main line. While neither are underpinned by MVCs, they will be generating revenue proportionate to their respective refinery’s utilization, which is highly ratable unlike a bet on refinery profits. 

As long as these refineries stay open, this revenue is most likely stable.

Offshore pipelines (56% of UFCF):

High-Quality Assets (~80% of offshore cash flow): Mars, Amberjack, Mattox, Poseidon, ~20% of Zydeco, Proteus/Endymion

Lower-Quality Assets (~20% of offshore cash flow): Auger, Eastern Corridor (Odyssey / Na Kika / Delta), Cleopatra

SHLX owns a collection of offshore corridor pipelines that transport crude oil from the Gulf of Mexico (GOM) deepwater to major crude trading hubs in Louisiana. SHLX’s offshore pipelines are essentially a capex-free, irreplaceable monopoly with annual price escalators and relatively low variability in production volumes.  Although greenfield projects can be very expensive, established platforms produce oil at very low marginal costs for the life of the project (30+ years in many cases).

Think of offshore pipelines as highways -- these were built pre-2000 with enough capacity to bring crude from 3-10 major production platforms to shore. SHLX owns substantially every oil highway to the Louisiana shore (or “central GOM”) and transports >70% of total Gulf of Mexico production. The central GOM is also home to most new major deepwater opportunities.

As a producer, these corridor pipelines are the only way to get to shore in the central/eastern GOM. The alternative is to build 100-300 miles of pipeline 3,000-7,000 feet underwater, costing on the order of a billion to construct (if you can get the permits). This puts SHLX in a huge position of power, which has resulted in highly favorable industry standards for pipeline operators, counter to what is typically seen in onshore pipelines:

1) De minimis capex to the pipeline operator: the producer pays all of the capex to connect the platform (wherever that might be) to the desired corridor pipeline – in contrast to the networks of gathering pipelines onshore (e.g. Permian) paid for by MLPs. In the rare instance growth capex is needed on a corridor pipeline to increase capacity, the producer responsible for exceeding previous capacity agrees to MVCs and frontloaded payments to guarantee a rate of return.

2) FERC+ price escalators: historical price increases equal to the FERC index, generally 3% / year. Better than FERC, this also has a floor of zero (whereas FERC can occasionally go negative on weak PPI years)

3) Life-of-lease contracts: producers agree to pay the full pipeline tariff on every drop of oil that comes out of the field (“life-of-lease commitment”), whether or not the physical oil is sent on that pipeline or separate pipe (to a different destination). Pipelines will not guarantee line space without this agreement, creating risk that the producer will not be able to get oil to shore for potentially disruptive periods of time.

What’s the catch? Due to life-of-lease contracts (as opposed to MVCs), revenues are levered to production volumes of the offshore projects. The common misconception is that current projects are a materially declining NPV of cash flows. To achieve a real multiple, many think you have to underwrite a long-tail of new multi-billion dollar greenfield projects, which is an implicit bet on a) oil prices, b) fruitfulness of new exploration activity and c) government regulation. That is enough to scare off a lot of investors.

We believe this is wrong – we think you can underwrite a substantial multiple based entirely on existing projects + upcoming projects that have passed Final Investment Decision (FID) by their sponsors (online in next 2-4 years). Based on studying production curves of platforms over time back to 1990 and talking to multiple industry experts, we have found the actual production curve is much flatter than one might expect.

The decay curve of a reserve with no incremental drilling is 10-20%, offset by two different types of projects:

1) Infield projects: there are a variety of projects to increase production in a given oil field, e.g. expansion projects (drilling more wells i.e. holes into an oil field already in production) and well enhancement projects (such as water injection to create more pressure in the reserve for oil extraction)

2) Tie-back opportunities: a producer may find a number of smaller oil reserves near the production platform. These are reserves that were too small to merit paying for its own standalone platform. However, these reserves can be tapped for low marginal cost “tying back” them back to the nearby production platform. Note that tie-back opportunities are generally not included in the FID…therefore platform cost is sunk and returns on these projects are purely incremental. 

Operators should pursue these projects in almost any oil-price environment. Marginal operating cost is ~$5-10 / bbl at any major oil production platform. Tiebacks and infield projects typically have <$20/bbl thresholds to surpass their IRR hurdles. We believe that as long as oil price stays around or above $20, these projects will be pursued as long as they are available.

The result of these projects is a much more palatable production curve. See below how production 10 years out is still 70-75% of initial peak volumes at an average platform. Platforms with rich incremental project opportunity can maintain production much longer than that; e.g. Mars / Ursa fields are 23 years old, yet are still producing at ~75% of peak oil levels (and growing).

Source: BOEM

Platforms used: Mars / Olympus / Ursa, Thunder Horse, Jack / St. Malo, Tahiti, Atlantis, Shenzi, Lucius, Mad Dog, Brutus, Bullwinkle, Holstein, Heidelberg

In the best corridors (i.e. along Mars and Amberjack), decay rates average out across the cycle of enhancement/tieback projects to 4-5% / year. In corridors with geologically worse oilfields  / less tieback opportunities (i.e. Odyssey and Auger), decay rates are faster at 9-10% per year. At the best sites, these additional projects can even boost production in excess of their initial peaks, e.g. Tahiti, Mars, Ursa, Atlantis and Bullwinkle. 

By layering projects past FID (using the above curve) on top of existing production, we believe SHLX offshore revenue will actually grow revenues from 2020 through 2023 with LSD declines through 2029 without a new major project ever passing FID. Our estimates through 2029 are below for reference. With unheroic assumptions, the NPV value of current + approved projects roughly equates to 8-8.5x of cash flow for the total offshore business. 

Also worth noting that the required oil price to clear IRR hurdle on greenfield GOM projects has dropped from ~$50+ / bbl in 2019 to $35-40 / bbl today. Drilling rig rates (~50% of the cost) have fallen ~50% since the covid downturn. At a $40 / bbl threshold, new projects do not seem a far stretch of the imagination.

Other pipelines (6% of UFCF):

Assets: 60% of Zydeco (Zydeco Mainline), LOCAP, Permian gas gathering pipelines

These consist of two crude oil pipelines and a network of gas gathering pipelines:

-Zydeco mainline is contracted with take-or-pay contracts, but is being competed down on price upon expiry due to competition from the new Bayou Bridge pipeline (and the reversal of Capline)

-LOCAP is a short crude pipeline between two Louisiana oil trading hubs without MVCs or contracted capacity

-The Permian assets are gathering pipelines that are levered to local Permian production, which are assets we have typically frowned upon (but are merely 1% of cash flow)

Valuation

We believe MLP investors do not appropriately discriminate between asset quality, capital intensity, counterparties, and longevity. Most MLPs we have seen own substantial assets with only a few years of cash flows left at materially above market rates, or require substantial ongoing capex to keep cash flows from declining. Despite that, most SOTP analyses treat all pipeline multiples the same. 

An illustration: median reported maintenance capex for large MLPs is 8.5% of EBITDA - which is substantially understated given many MLPs classify certain sustaining capex as growth capex (e.g. building gathering pipelines). Meanwhile, SHLX spends <4% of EBITDA on maint capex. Notably the offshore assets have exceptionally low maint capex requirements - for example we believe that Mars maint capex has averaged <$1mm yearly over the last 20 years (for reference, Mars did $188mm of EBITDA in 2019). This is one of many reasons we believe exceptional assets like Mars should command a significant premium to the typical onshore crude pipeline owned by most MLPs.

We value SHLX on SOTP based on our estimates of unlevered FCF:

1) Refined product pipelines (18% of UFCF) - 14x | ~7% yield: comparable to multiples achieved in PE transactions and sum of parts valuations for other MLPs. 

2) Refinery-based assets (20% of UFCF) - 12x | ~8% yield: highly ratable revenues, like refined pipelines. Long dated take-or-pays underpin value. 

3) High-quality offshore assets (46% of UFCF) - 11.5x | 8%+ yield : we delineate the higher quality offshore pipelines from lower quality ones, based on a) average project decay rates in local basins and b) runway of FID’ed projects. Our NPV value of existing + FID’ed projects on high quality pipelines is 9-10x FCF. This is calculated with unheroic assumptions: 2% price escalators matching cost inflation, 5-10% annual volume declines in existing projects, 8% discount rate. We believe the remaining stub of valuation is an acknowledgement that future FIDs are very feasible, especially with project costs down 30% from 6 months ago due to low rig rates. Transaction comps are messy, as transactions were sold either in asset bundles (GEL) or artificially-low mult dropdowns (SHLX / BPMP). However, we think the implied multiples for the higher quality offshore assets in the 2015 Enterprise / GEL transaction were at minimum 10x given many of the other assets were declining. 

4) Low-quality offshore assets (10% of UFCF) - 7x | 14%+ yield: The NPV value of existing projects (unfortunately no new major FID’s in the hopper on lower quality assets) at 8-10% annual volume declines is roughly 6-7x FCF. 

5) Other pipelines (6% of UFCF) - 6.5x | 15%+ yield: we place an unassuming multiple and move on. We do however note that sell-side places 8x+ multiples on these assets.

Shell parent owns SHLX preferred stock, which converts at $23.63 / share at the option of Shell parent beginning in 2022. We value the preferred stock at face for conservatism, although this likely overstates the value given a) the preferred pays a measly 4% dividend and b) the “option value” is way out of the money.

Note: we treat JV LFCF as UFCF so we can exclude JV debt in capitalization, consistent.with how SHLX reports adj. EBITDA.

We arrive at a blended 8.5%+ UFCF yield, yielding 43% upside at this price. We note that our bull case at 77% upside is still at a discount to pre-covid levels. Our bear case assumes a second lockdown as brutal as the first, which shows valuation support (absent another leverage implosion at closed-end MLP funds) above $10 / share. We also note that none of our valuation cases bake-in any future value creation through dropdowns, which Shell has a long history of doing at favorable multiples.

Risks:

-Another lockdown halting driving for 3+ months, flowing through the chain to refinery production / oil production

-The broader MLP space trading down on any macro news e.g. DAPL ruling, dragging down SHLX along with it. This could also be exacerbated if any of the major MLP funds relever and get caught offsides again

-Joe Biden has adopted strong rhetoric against US oil production during the Democratic primaries, which we believe is more bark than bite. To play pessimist, we think the worst reasonable outcome is a temporary halt + longer approval process on new government approvals for greenfield offshore projects. We view it highly improbable that existing projects are impeded - which leaves the majority of our valuation safe. We note that all four attempts by Obama to halt deepwater oil activity after the Macondo disaster were overturned by courts. Nevertheless, most flavors of Biden regulatory action on offshore activity introduces headline risk and some marginal risk to value.

Appendix

Figure 1: Asset Breakout

Figure 2: Historical Offshore Production Curves

Source: BOEM

Catalyst

-Demonstration of ratable earnings in a shaky environment during Q2, Q3 and beyond

-More reassuring dividend commentary

-Time as funds are able to do the work on a complex set of assets for the first time since the IDR elimination