Description

Thesis: MPC is the largest refiner in the country and at $29 we can implicitly buy those refining assets for just $3.4-5.6bn, or 1.4x to 2.3x my estimate of normalized Ebitda, with proceeds from the sale of Speedway being used to potentially repurchase ~25-45% of the total market cap at that depressed valuation near cyclical lows. The low-end of the buyback range assumes pro-forma net debt of zero, while the high end would assume just 1x leverage on normalized Ebitda. Due to its special situation status, MPC is one of the most out of favor stocks in one of the most out of favor sectors of the market. Divergent views on the use of proceeds from the recently announced sale of Speedway has created a fractured investor base with neither side willing to deploy capital prior to the 1Q21 expected close. As management clarifies its plan for the $16.5bn+ of proceeds over the coming months, potentially on the 3Q call on Nov 2nd, it should create a clearing event for the stock. 

Based on recent management comments, the likely course of action will be a massive share repurchase. That will disappoint the simplification crowd, but the sheer size of the share repurchase should push the stock higher over the coming months. If management surprises the market and buys-in the minority portion of MPLX, the resulting simplification of the corporate structure should finally enable a re-rating. In both scenarios I think the stock is worth $38-39 (+30-35%) in the base case on normalized mid-cycle refining earnings, and potentially 2.0-2.5x over time in a bull case (pre-covid levels would be 2x alone). MPC should also benefit as a re-opening play (back-to-work miles driven and airline traffic). Of course if lockdowns re-emerge and the duration to refined product demand/inventory normalization is pushed out, the stock will take longer to work and the company will burn more cash, but at the current price the risk-reward seems quite compelling.

High-Level Math & Valuation: 

• MPC currently has an EV of $29.5bn on a stand-alone basis (excluding non-recourse debt from MPLX that it consolidates). After subtracting the value of its 647m MPLX common units worth $11.0bn (at $17.00), and the $16.5bn+ of after-tax proceeds from the sale of the Speedway convenience store business to 7-11, the pro-forma company (the largest refining system in the US) is being valued by the market at just $3.4bn. 

• That also assumes ~$1.35bn of cash burn at MPC from 3Q20 until Speedway closes in 1Q21 (including dividends) using near-trough crack spreads and low utilization. Even if operating cash burn ends up being a bit higher, I have conservatively excluded the potential for higher than expected after-tax cash proceeds from Speedway and/or over $1bn in one-time CARES Act related tax refunds in 1H21. 

• To be even more conservative, and applying a 20% discount on the FMV of MPLX (which is already being discounted by the market at close to an 18% levered FCF yield), then the pro-forma valuation of core MPC would be ~$5.6bn. 

• At my conservative mid-cycle estimate of ~$2.4bn of Ebitda, pro-forma MPC is trading at just ~1.4-2.3x Ebitda. That seems too cheap no matter your view on the corporate structure or the terminal value of the refining industry. 

• Alternatively, at a 5x multiple on the $2.4bn of mid-cycle pro-forma MPC Ebitda, MPLX equity is being valued at a 80% discount to current FMV, implying a 5.1x midstream multiple and a ~16% unlevered FCF yield.

• MPC currently has ~$10.6bn of parent net debt (excluding MPLX). Proceeds from Speedway will be at least $16.5bn, possibly more. That leaves about $4.6bn of excess cash to deploy (on a $18.9bn mkt cap) if MPC decides to be conservative and leave its balance sheet with zero net debt until the product market normalizes. That would amount to repurchase capacity of ~24% of the company.

• Management has indicated it would like to leave some leverage on the company. With $2.4bn of normalized Ebitda and ~$1.8bn of distributions from MPLX (assumed as Ebitda by the ratings agencies), and levering that $4.2bn of total Ebitda just 1x would produce an incremental ~22% of repurchase capacity.

• Refiners have historically traded for about 4-6x Ebitda. These low multiples reflect their earnings volatility, capex intensity, and limited growth opportunities.
• At a 5.0x Ebitda multiple on normalized Ebitda of $2.4bn, I see MPC worth $39, which doesn’t include any value creation from buying back stock with the Speedway proceeds before the cycle turns, and assumes a 20% discount on MPLX.

• Upside could be >$70 if refining fundamentals revert to above mid-cycle earnings. With fundamental downside to ~$21 assuming a new-normal mid-cycle earnings power is materially below the prior cycle, the risk-reward proposition skews favorably with 1.3x more upside than downside under the base and bear case, and >5.5x more upside than downside in the bull-bear scenario. 

• Alternatively, if we count MPLX’s distribution to MPC as Ebitda and capitalize that at 5x along with the refining earnings (despite the distribution being cash flow vs Ebitda), and MPC would still be worth $39. 

• Please see the end of the report for more details into my normalized Ebitda assumptions, but note that excluding Speedway, pro-forma MPC (exc Speedway) generated $4.8bn of Ebitda in 2019.

• Despite the volatile earnings characteristics, refining businesses generate decent free cash flow across the cycle. With investors and management teams demanding higher hurdle rates on projects, future capex should be lower and through-cycle FCF could be higher. 

• Using $750m of maintenance capex, a conservative estimate for growth capex of $1.0bn, and MPC should generate over $4.00 per share of mid-cycle unlevered FCF (assuming zero net debt), or a ~14% yield at today’s price. In a bull case scenario, MPC could generate ~$11 of unlevered FCF, close to 38% of the current enterprise value. In the bear case MPC will muddle along around free cash flow breakeven with $1bn of growth capex it could cut.

Why the discounted valuation? 

The simplest reason is due to poor fundamentals, cash burn, and the markets disdain for anything energy (oil and gas) related. The market is unwilling to look out longer than 6-12 months to what level of earnings these businesses could and are likely to generate once demand returns to pre-covid levels and inventories normalize. Most investors don’t expect refining fundamentals to completely normalize until 2022/23E, and no one appears willing to play that time-arbitrage with the backdrop of a secular declining business. The prevailing belief is that oil and refined product consumption has peaked. But if the economy is to recover to pre-covid levels, it will need oil and refined products to a greater degree than previously. The longer-term outlook is anyone’s guess, but at current valuation there is no terminal value being embedded anyways.

The second reason is MPC’s history and consistent SOTP discount. MPC was spun out of MRO in 2011 with a diverse set of assets including refining, midstream, and retail (c-stores) and has never been given the appropriate value on a SOTP basis. It has been the subject of activist campaigns since 2012, with multiple big-name firms (JANA, Elliott, Corvex) agitating at various points in time and pushing management, under former CEO Gary Heminger, to unlock value in creative ways. MPC often did many of the things those firms wanted to help remove the SOTP discount, from the creation of MPLX with its midstream assets in 2012, to the simplification and monetization of its general partner interest in MPLX in early 2018, to the sale of Speedway in 2020. All the while MPC has repurchased shares using excess capital to take advantage of the valuation gap. However, every time it has taken steps to remove its SOTP discount, it would then go and do something that investors didn’t want that either destroyed value and/or added more complexity. This would include the top-of-the-cycle purchases of MarkWest by MPLX in 2015, and ANDV by MPC in 2018.

The recent strategic reviews of Speedway and Midstream also failed to garner much confidence in management’s willingness to be bold and simplify the business once and for all. The original decision to spin-off Speedway was met with skepticism because it would have resulted in pro-forma MPC having too much leverage relative to its volatile earnings. And the decision to keep the current structure of the MPC/MPLX relationship in early 2020 was anti-climactic. That is why despite the announced sale of Speedway simplifying the SOTP (from three to two parts) at a valuation in line or higher than most analysts’ expectations, the market appears to be skeptical over whether a) the Speedway sale will close (and whether it will close on time), and b) what the use of the Speedway proceeds will be when and if it does.

Will Speedway Close? 

The risk of the $21bn Speedway sale not closing seems to be quite low. On the financing front, Seven & I (7-11) is well capitalized in a net cash position and has access to cheap financing being domiciled in Japan. Net leverage post-close should be <2.5x, and that is before the $5bn of sale-leaseback proceeds and synergies from which it expects to drive Ebitda up by ~1/3. 

On the regulatory front, the US c-store landscape is very fragmented. Despite 7-11’s current #1 position in the US, post-close it will still have a <10% share of the overall US market. The two chains appear quite complementary and there appears to be limited overlap, so divestitures should be small and driven by local market density. There should be ample demand for those divested assets at good multiples. My understanding of the asset purchase agreement (please double check) is that 7-11 has the option to break the deal if divestitures are >$400m of Ebitda (of the $1.5bn). Both 7-11 and Speedway have significant experience in c-store M&A and dealing with regulators so there shouldn’t be any reason why this process should take materially longer than anticipated (1Q21 close, May 2, 2021 outside date) or require more divestitures than both parties believed. Per discussions with management, both companies did their own analysis and came to the same conclusion over divestitures and timing. The FTC is reportedly doing a quick 30-day review and may require a second review which is how the 5-7 month time frame (from Aug 3 announcement) was calculated. Based on management conversations, it seems like the process could be quicker than expected which would be a positive surprise.

Use of Proceeds The Biggest Focus Point: 

The use of proceeds is the biggest subject of debate for the stock. There are two shareholder bases with differing opinions on the appropriate route to create value going forward. The buyback investor base wants all proceeds to be used to de-lever and repurchase shares, while the MPLX buy-in investor base wants MPC to put an end to the complex corporate structure once and for all by buying-in the remaining 39% of MPLX that it doesn’t already own. I believe these divergent opinions coupled with management not giving a definitive answer (given the rapidly changing fundamental landscape), is the primary driver for MPC lagging the S&P, refining peers, and the overall energy sector since the deal was announced. Both investor bases fear the other scenario, and neither is willing to deploy capital prior to a firmer commitment on the use of proceeds, creating an orphaned special situation stock.

It is my belief, based on recent management comments, that what WILL happen is that most, if not all, proceeds will be used to repurchase shares while de-leveraging. New CEO Mike Hennigan clearly stated on the last earnings call that this is a return of capital business and that they did not believe it made sense to buy-in MPLX. The market appears to be confusing Hennigan’s commitment to explore the best way to return capital (tender, open market repos, special dividend) with a more open-ended review that could include an MPLX buy-in. 

Mike Hennigan, Aug 3, 2020:

• “I believe this is a return of capital business and a substantial estimated after-tax proceeds of approximately $16.5 billion enables us to both strengthen the balance sheet and return capital to shareholders.”

•  “On use of proceeds and capital, I’ll make a few comments and then I’ll let Don also touch on the topic. First of all, I’ve stated previously many times and in prepared remarks that I believe this is a return of capital business, and I’ve made a statement that we have three key initiatives going on, optimizing the portfolio and evaluating assets for free cash flow generation, improving our commercial performance and reducing costs. And I think all of these are aimed at returning more capital to shareholders. And this transaction is no different. We expect two primary uses of proceeds. One is to defend our investment grade profile, and Don’s going to talk about that a little more, and then returning capital to shareholders.

•  “And with respect to return of capital, we know this is an important component of value realization, and we are committed to it. The issue we have right now, though, if there wasn’t a global pandemic and a global recession in progress and we had a much clearer picture of refining supply and demand going forward, it’d be much easier to give a little bit more detail. But between now and the closing date, were going to continue to evaluate the most efficient and effective means to accomplish it.”

• On whether to buy-in MPLX…”And the bottom line is we came to the conclusion that we didn’t want to buy in a cash stream that we were already getting with respect to the RLFD (refining logistics and fuel distribution). Marathon has the unique position where we’re receiving about $1.8 billion in distributions from MPLX, and RLFD is about $1.4 billion of Ebitda. So, again, I’m saying we came to that conclusion of why would we buy in that cash stream when we’re getting it essentially via distributions and instead use that capital, which would be north of $10bn, in a return to MPC shareholders.”

•  “So, at this point, we do recognize and we are frustrated at the level of equity value that MPLX has. At the same time, we felt – or continue to feel that the amount of capital that would be needed to do that at this time makes more sense to return to shareholders as opposed to not…And as I mentioned and Don commented as well is were going to be doing a lot of analysis between now and the close to look at what’s the best use of proceeds. But the highlight priorities are the balance sheet and returning capital to shareholders.”

I view the decision to return capital to shareholders (most likely via buybacks) as the best use of proceeds as we are closer to the bottom of the cycle than the top, and by repurchasing MPC shares cheap, management is implicitly purchasing MPLX shares for even cheaper (80% discount to current equity FMV). There are other ways to simplify the corporate structure down the road once the cycle normalizes. In the meantime, the sheer magnitude of the share repurchases (24-46% of total float) should force the share price higher even as part of the investor base churns initially. Based on recent sell-side notes, it appears a larger percentage of investors believe this is the route management will take, but there has been a growing chorus of proponents of an MPLX buy-in.

There have been a number of scenarios for how MPC could simplify the corporate structure through the years. Assuming management executes a massive share repurchase, the one that seems to make the most sense from a high level (whether actually possible or not positive) when refining and oil/gas fundamentals improve is the spinning off of some or all of its ownership of MPLX into a newly traded C-Corp (using partial spin to de-lever). It could then merge MPLX with the C-Corp, helping to garner more fund flows and a higher share price, while also bringing its ownership below the 50% threshold required for financial statement consolidation. With MPLX being valued at an 80% discount within MPC today, there appears to be no good reason to address this issue today. The better use of capital would be to implicitly repurchase those shares at the 80% discount and deal with the overall corporate structure down the road. Kick the can until fundamentals improve and buy back stock until then.

What if MPC buys-in MPLX? 

Because a larger percentage of the shareholder base expects (and I believe wants) MPC to repurchase a large percentage of shares, the big risk is that management once again pulls the rug out and buys-in MPLX. If this happens the stock will go down initially, but I think this is already being mostly priced in (famous last words), and is actually not all that bad of an outcome. If refining fundamentals normalize, the combined company would still have significant upside. The MPLX buy-in investor base wants the corporate structure simplified to eliminate the inter-company refining logistics and fuel distribution costs that make MPC’s refining business appear higher cost and lower margin. Many also view MPLX as a captive MLP that is a relic of the financial engineering days, especially with 90% of its Logistics & Storage Ebitda coming from MPC. With MPLX (and all MLP’s) trading at depressed multiples, the belief is that now is the best and cheapest time to do this, especially given MPLX’s commitment to reduce capex and generate significant FCF in 2021. But as reiterated by Mike Hennigan above, the view from MPC when it concluded its midstream review was that it wasn’t worth spending the incremental capital to get full control of MPLX because it was already receiving $1.8bn in distributions.

• If MPC buys-in the remaining 39% of MPLX it doesn’t already own at $18.50 (~9% premium to $17.00 and roughly the average stock price since May), it would only cost MPC ~$7.6bn. It would imply a take-out Ebitda multiple of ~8.0x (about the same multiple as prior drop-downs) and would require MPC to take on MPLX’s debt load. 

• Pro-forma leverage would be ~3.3x in my base case using the $2.4bn of normalized legacy MPC Ebitda and it would imply pro-forma MPC/MPLX trading at a ~5.8x multiple. Still quite cheap given ~70% of the Ebitda would be coming from the more stable MPLX “midstream” earnings base. 

• The merger of MPC and MPLX would create an entity that would be generating about $7.6bn of Ebitda and ~$4.75 of pro-forma FCF at that 3x leverage level, or a 16% levered FCF yield. That free cash flow would enable MPC to de-lever by ~$1.6bn annually, or ~0.2x (after paying dividends). 

• At a reasonable 6.5x multiple the stock would be worth ~$38, and in a bull-case it could be worth over $100. Even in a prolongued bear case the pro-forma company could generate positive FCF.

Mid-Cycle Ebitda Assumptions: 

For a generalist investor, estimating the mid-cycle earnings power for a refiner is hard enough given the volatility in oil/refined product prices, crude differentials, and supply/demand, but is made even tougher for MPC because of the changes in its refining business over the last few years. 

On Oct 1, 2018, MPC completed the purchase of ANDV, which added 60% more refining capacity, mostly on the West Coast. In its quest to create value and simplify its business over the last several years, MPC also sold refining logistics and fuels distribution assets to MPLX – boosting Ebitda at MPLX while increasing fixed costs at MPC. These changes make it difficult to look at the refining business’ earnings over time and eyeball what normalized Ebitda should be. 

Below I break down the key assumptions for estimating my base, bear, and bull cases for pro-forma MPC’s earnings power:

Refined Product Volume: 

• MPC is the largest refiner in the country with close to 3.1mbpd of capacity. It recently shut down its legacy ANDV refineries in Martinez, CA and Gallup, NM accounting for ~6% of total capacity. Historically MPC has run its refining system at a 95-98% overall operating rate. To account for the closures of Martinez and Gallup, I assume a 90% normalized operating rate on current capacity which equates to about ~1.0bn barrels of refined product output per year. 

• The system is currently running at 70-75% of capacity due to excess inventory in the market caused by Covid. However, once this inventory is burned off there is typically not much annual volatility in operating rates. The timing of multi-year maintenance turnaround schedules is usually the primary driver of any volatility. The industry normally operates at high utilization rates to maximize fixed cost absorption, while near-term supply-demand imbalances are generally reflected in crack spreads.

Gross Margin:

• The crack spread is the key assumption in  forecasting normalized earnings power for a refiner. It is the differential between the input price of crude and the sale of the refined products. Utilizing MPC’s current methodology for calculating the indicated crack spread in each region it operates (Mid-Con, Gulf Coast, and West Coast), and back-testing that to 2006, I believe May 2011 to Nov 2019 represents a decent analog for what a mid-cycle earnings period should look like (see second chart below). 

• During that time period, the blended crack indicator averaged ~$15.75/barrel over all rolling 12-month periods (range of $13.50-20). Adjusting for the cost of RIN’s and that equates to a ~$14.50/barrel blended crack spread. If the “new normal” is at the low-end of the prior range, I estimate a ~$13.25 blended crack, and the bull case could be $17+. Spreads are currently at trough levels not seen since 2008/09 as shown in the first chart.

• Besides the crack spread, the two other primary factors impacting refining gross margin are the sweet-sour crude differentials and the overall capture rate. The sweet-sour differentials represent the delta between the delivered cost of MPC’s crude feedstock and the crude oil reference price (ie WTI) used in the crack indicator. I conservatively estimate $0.50/barrel of margin but that has at times blown out to several dollars per barrel depending on intra-basin bottlenecks and spreads. If oil ever recovers and producers once again lose their discipline, MPC should stand to benefit via crude differentials between basins. 

• The capture rate represents how efficient MPC is at operating its refining system relative to the market reference crack spread and crude differentials. MPC’s goal is to achieve a 90% capture rate, which it was able to do in 2019 despite being the first year post-ANDV merger. It is possible that as MPC optimizes its system, leverages its scale, and achieves marketing synergies it could achieve a higher capture rate but I think using 90% is fair for a base case, 85% for a bear case, and 95% for the bull case.

• Putting it all together and I think a $13.50 gross margin per barrel ($13.6bn of gross profit on ~1bn+ barrels) is a reasonable/conservative estimate for mid-cycle normalized earnings power of the refining business. To put that into perspective, 2019 was the first full year post-ANDV, included an abnormally weak 1Q19, and gross margin was $14.01. That compares to 2Q/3Q20E trough margins that will be ~$7.15. MPC’s refining gross margins averaged $13.75 from 2013-18, and those legacy margins didn’t include ANDV’s west coast operations which typically have higher margins than the Gulf and Mid-con. Bears will say that margins will be lower going forward due to reduced demand and narrower crude differentials, but there is also a possibility that more high-cost refining capacity shuts down across the US as has happened recently.

 Operating Expenses:

• I estimate refining operating costs to be ~$6bn at a normal utilization rate. MPC is currently cutting operating costs and expects a decent chunk of the $750m of costs to be taken out in 2020 to be permanent. Operating costs in 2019 were $6.4bn, but that was before costs cuts, full realization of ANDV synergies, and before the closure of the high-cost Martinez and Gallup refineries. I think $6bn (or $6 per barrel) is a decent run-rate when utilization finally returns to normal.

• MPC pays MPLX and other third-parties fees to transport, store, distribute and market both crude oil and refined products. These fees are mostly fixed in nature as they are typically based on volumes with minimum volume commitments and should reach a ~$5.2bn run-rate once MPC gets back to normal utilization. Of the $5.2bn in total distribution costs, ~$3.4bn are paid to MPLX (>90% of MPLX’s L&S segment earnings are from MPC). A large chunk of these costs were dropped down to MPLX in a bid to create value by garnering a higher multiple within MPLX. However, that has not panned out as most investors now view these distribution costs as contributing to MPC’s higher operating costs and lower margins than peers.

Corporate & Other:

• After Speedway is sold, MPC will retain most of the run-rate corporate costs, though it is possible there is some cost cutting that can happen as a smaller organization. 

• MPC will also retain the direct dealer wholesale fuel distribution business with ~1,075 locations (under the ARCO brand). This business generates ~$450m of Ebitda, but is volatile. 

• MPC also has midstream assets that are not owned by MPLX. The most valuable is a 25% interest in the recently commissioned $2.7bn Gray Oak crude pipeline from the Permian to the Gulf. Along with other smaller pipelines, including the expected reversal of the Capline pipeline, and I estimate roughly $175m of proportionate Ebitda to MPC.

• Corporate costs plus MPC’s other businesses/assets should net close to zero, leaving the refining business as all that really matters for pro-forma MPC.

Total Pro-forma Normalized Ebitda:

• Pro-forma post-Speedway MPC (exc MPLX) should generate at least ~$2.4bn of mid-cycle Ebitda. That compares to 2019 exc Speedway Ebitda of ~$4.8bn, so not a heroic assumption. 

• In periods of bullish refining fundamentals, MPC could generate over $6bn of Ebitda. In the bear case, where the new normal is materially lower than the last decade, MPC wouldn’t generate much Ebitda at all through the cycle. I think the probability of the bear case over a prolonged period is quite low, and mostly being priced in today.

• There is some upside to the refining business earnings power not included in my base case. This includes upside from recent capex projects underway but not yet completed (STAR), incremental Ebitda upside from the future conversion of the Martinez and Dickinson refineries to renewable diesel, and upside from the Speedway/7-11 fuels distribution agreement. It also doesn’t include any market related upside from extraneous factors such as IMO 2020, which up until recently was THE refining thesis.

Risks: 

The key risks are that 7-11 backs out of the Speedway deal or there is a lower for longer refining cycle in which MPC burns more cash than expected after it sells Speedway. If MPC goes the share buyback route, the  core pro-forma business will be quite volatile given the lack of Speedway earnings. The market may not take into account the offsetting stability of the $1.8bn of distributions received from MPLX and will thus put a lower than peer multiple on the stock. The election represents market risk as a Biden victory will be perceived as negative for the oil/gas sector, and MPC reports the day prior. Lastly, if MPLX earnings get hit due to lower G&P or pipeline volumes, its distribution could be perceived as at risk (there is plenty of coverage once it cuts capex in 2021+).

Catalyst

The two hard catalysts that should drive MPC’s share price higher are 1) the close of the Speedway sale and 2) an announcement providing more clarity on the use of proceeds, potentially coming in less than two weeks (but likely not until 1Q21E). The primary soft catalyst is the gradual bottoming and subsequent improvement in refining fundamentals. Some analysts currently believe that the Speedway close and announcement on the use of proceeds will be a sell-the-news event either way because of the churn between investor bases. That could be true but that will provide the final near-term clearing event needed to attract more flows. One incremental catalyst could be the sale of MPLX’s G&P business at a multiple higher than expected. At a 7-8x sale multiple, that would bring MPLX’s net debt to 2.7-3.2x, which also would optically reduce the consolidated debt load of MPC.