Description

Marathon Petroleum (MPC) is a collection of refining, midstream, and gas station/convenience store assets trading at a discount to the sum-of-the-parts. Catalysts within the next 6 months should eliminate the discount.

Later in this write-up I’ll give details of my sum of the parts using economic earnings, but the high level is that you’re paying 7.5x EBITDA for something where hard catalysts exist to realize >10x EBITDA for the vast majority of the cash flows. I get to a SOTP value of $65/share once these catalysts play out.

It’s important to understand MPC’s structure and accounting treatment because it partially speaks to the mispricing. MPC owns the GP and 26% of the LP units of MPLX, an MLP containing gathering & processing assets mostly in the Marcellus and logistics assets associated with MPC’s refineries. MPC reports MPLX on a consolidated basis. If you pull up MPC in your data provider, you will most likely see an Enterprise Value calculated including the balance sheet and minority interests of MPLX and the combined financial statements of MPC and MPLX. Segments are reported as Refining & Marketing, Midstream, and Speedway (c-stores).

On an LTM basis, it looks like MPC trades at 9.6x EBITDA, with over half of that EBITDA coming from Refining & Marketing.

A better way to look at MPC is by looking at the parent standalone. Below is FY 2016 EBITDA by segment (with corporate overhead allocated to the segments). Importantly, I broke out the Refining & Marketing segments into the components.

FY 2016 EBITDA:

Refining - $1,065

Marketing - $1,400

Speedway - $887

Midstream (outside of MPLX) - $264

GP IDR cash flows - $172

Other - $24

Total - $3,812

Thus on a parent EV of $28.5b (backing out current MPLX units), MPC trades at 7.5x FY’16 EBITDA.

So far this is just fun with numbers, but over the next 6 months the following will happen:

1. Marketing assets with qualifying EBITDA of $1.4b will be dropped into MPLX at 7-9x EBITDA. The first drop of $250mm already happened at 8x, and the remaining $1.15b will probably also happen at 8x. MPC receives units and cash as consideration.

2. The IDR’s will be exchanged for MPLX units at a range of 15-20x. Note that after the drops the IDR cash flows will be over $600mm.

3. Speedway will most likely be spun out to MPC shareholders.  Comps trade at 10x EBITDA.  

Management is taking these aggressive actions because of pressure from shareholders including Elliott. Management seems to be fully on board with all of the proposals. I’m using management’s timeline and proposed multiples for the above transactions. In recent presentations you can find tons of detail on the timeline of these events. You can also find the Elliott presentation from last November, which lays out their more aggressive assumptions.

There are three major areas where things could go wrong:

1) At the end of these transactions MPC will own $19.5b worth of MPLX units (worth $37.50 / MPC share). You can short MPLX or an MLP index to somewhat mitigate this risk, but it’s still a huge portion of the value.

2) The spin-off of Speedway isn’t assured. Management is completing it’s assessment in Q3 and public commentary seems very supportive of the spin, but anything is possible.

3) $600mm of the EBITDA being dropped into MPLX is associated with the wholesale refined products business. Management is seeking a PLR for this asset but has also said it’s not entirely necessary. Precedent exists for this type of business to be dropped into an MLP. CST (spun from Valero) did it with their wholesale fuels business, and others have as well. MPC’s wholesale business is a little bit different because it is “refinery to rack” instead of “rack to retail,” but it’s the same otherwise. The grey area of qualifying income is where it is closer to the end consumer not further back.

Below I’ll give more details on each segment’s business and valuation.

Refining

MPC owns a balanced set of refineries with 1.8mm barrels/day of throughput and a NCI of 11. Since 2000, a number of factors have turned U.S. based refineries into OK businesses from bad businesses. First, we’ve seen a rationalization from 300 refineries in the 80s to about 120 today. Second, more stringent environmental standards have reduced flexibility in operations, which lowers the sustainable run rate of refineries. Third, and most importantly, U.S based refiners gained a nice gas cost advantage as well as access to much more domestically sourced crude. The gas cost advantage is significant: as an example, if U.S. gas is $2.60 and European gas is $4.60, a refiner of MPC’s size would be advantaged by almost $500mm pre-tax!

Refining is still a cyclical business though. In 2011 and 2012, helped by huge differentials (domestically produced crude was cheap because more was being produced than could be taken away by pipes), MPC earned $10/bbl of operating income vs. $3 in 2016.

So to value the refining business I apply a multiple of free cash flow to mid-cycle. The key components of a refiner’s margins can be thought of as the difference between product prices and crude prices (“crack spreads”) and the difference between the benchmark crude and whatever advantaged crude can be sourced more cheaply.

My determination of mid-cycle is as follows:

$1,654mm of EBITDA equates to $842mm of EBIT and about $550mm of NOPAT, which I value at $7.1b. As a sanity check, my EBIT number isn’t too far above just the advantage from cheap domestic gas.

Speedway

Speedway consists of over 2,700 gas station / c-stores where 60% of gross profits come from the c-store. C-stores exhibit little to no cyclicality and are deserving of a high multiple. MPC is a decent operator of these assets, and the locations are large, well-maintained, and highly profitable.

This chart from MPC shows Speedway vs. others in FY’15:

Fuel margins are the most cyclical part of the business, and using 16 cents as mid-cycle I get EBITDA of $900mm and NOPAT of $480. I value Speedway at $7.9b, which is just under 9x normal EBITDA.

Midstream / MPLX

I’m going to spend less time on these assets because I’m not too far off how the market values MPLX. It’s worth noting that at the end of the drop-downs this will be a 50/50 mix of logistics & storage assets associated with MPC refineries and of gathering & processing assets in mostly the Marcellus. Both of these assets are well positioned. On the gathering & processing side, there is little commodity exposure - 92% is fee based. The balance sheet is IG, and the yield is reasonable vs. other MLPs.

SOTP

Putting all of it together gets us to about $65 of fair value.

Key assumptions for the mid case:

- I’m using current MPLX prices for the unit value as well as for the value at the time of the drops.

- $1.15b of qualifying EBITDA is dropped at 8x. 146mm units are received as well as $4.16b cash taxed at 20%. Management guided to 50/50 cash, stock and the first drop of $250mm was done with the stock at $38 and with $1.5b cash.

- IDR cash flow is 600mm after the drops. It is exchanged at 17.5x .

- Parent cash and debt are as of 3/31/17.

Catalyst

-Drop downs of MLP qualifying income
-IDR exchange
-Spin-off of Speedway