MPLX (MPLX)

Sponsor Risk at the Coverage Floor

MPLX is the clearest refiner-sponsored gathering and logistics MLP among the large partnerships, and the two cash figures it files point in opposite directions. Distributable cash flow, the partnership's own measure of cash available to unitholders and nothing to do with a discounted cash flow, was $5,791 million in FY2025 on $7,017 million of adjusted EBITDA. That covered the distribution 1.4 times for the year and 1.3 times in the fourth quarter: covered, and sitting on the comfort floor rather than above it. Free cash flow after all capital spending and distributions was negative $2,991 million over the same year. Leverage was 3.7× on face-value total debt divided by LTM adjusted EBITDA, with management supporting a range around 4.0×. Market capitalisation was roughly $57.6 billion on 10 June 2026.

Marathon Petroleum owned 63.7% of the common units and 100% of the general partner at year-end 2025, which puts it on both sides of the payout. It is the partnership's main commercial counterparty and it collected roughly $2.6 billion of the $4.1 billion distributed for 2025. Dropdown acquisitions and intercompany flows are part of how the structure is meant to work, not one-off events. Underwriting here means reading sponsor alignment and conflict alongside coverage and leverage, not instead of them.

Business Overview

MPLX operates gathering, processing, and logistics assets tied to Marathon's refining and upstream footprint, plus third-party volumes where contracts allow. Adjusted EBITDA of $7,017 million places MPLX mid-sized within the six-name set, below Kinder Morgan's $8,391 million and further below Enterprise and Energy Transfer.

The general partner arrangement is simpler than the label suggests. Marathon holds 100% of MPLX GP LLC, but the incentive distribution rights were bought out in February 2018, when Marathon exchanged its GP economic interests for 275 million newly issued common units. No IDR tier now skims the top of each distribution increase, so Marathon's economics run through unit ownership. Control does not: unitholders do not elect the board, and the sponsor appoints it.

MPLX publishes no consolidated fee-based share of EBITDA, and that matters less than it sounds. Fee-based is a wide category. It covers a take-or-pay contract, where the shipper pays whether or not it ships, and it covers a plain per-barrel tariff that earns nothing if volumes stop. A partnership quoting 90% fee-based may be describing either, and take-or-pay does not remove volume risk so much as convert it into the shipper's credit risk. The useful questions here are more specific: how much throughput originates with Marathon, and whether the minimum volume commitments bind on the assets that would actually go quiet in a refining downturn. Both sit in the contract and related-party footnotes rather than in a headline percentage.

How the Economics Work

Distributable cash flow of $5,791 million is the figure the distribution is set against. At 1.4× coverage, implied distributions were roughly $4,136 million ($5,791 million ÷ 1.4), which cross-checks against the $4.0660 declared per unit on about 1,015 million units. That is a yield of about 7.2% on the ~$57.6 billion cap, at the upper end of the large-cap MLP band of roughly 5–8%, consistent with tighter coverage than Enterprise at 1.7× or Energy Transfer at roughly 1.80×.

The Q4 print of 1.3× is not noise. MPLX raised the quarterly distribution 12.5% in October 2025, to $1.0765 per unit, so 2025 carries two quarters at the old rate and two at the new. Hold the new rate for four full quarters and it costs about $4,371 million, which is coverage of roughly 1.3× on unchanged distributable cash flow. The 1.4× is a blend that does not repeat unless cash flow grows into it.

Leverage at 3.7× uses face-value total debt, not net debt with hybrid credits. Cross-peer leverage comparisons require matching definitions: Enterprise adjusts hybrids; Enbridge uses debt-to-EBITDA; Energy Transfer cites a qualitative range. MPLX at 3.7× is inside its supported ~4.0× range but closer to the ceiling than Enterprise at 3.3×.

Dropdowns from Marathon arrive in lumps, adding both EBITDA and distributable cash flow at once, and the market usually capitalises them at partnership multiples near the 8–10× screening band. Whether a drop improves coverage depends on the price paid, how it was financed, and what the acquired assets cost to keep running, not on the EBITDA added.

Where the Coverage Number Is Movable

Distributable cash flow deducts maintenance capital and ignores growth capital, and no accounting standard draws the line between them. The company draws it, and no auditor tests where it fell. MPLX spent $252 million on maintenance in 2025 against $2,037 million on growth, so 89% of the capital budget sat outside the metric that sets the payout. Call a compressor rebuild growth rather than maintenance and distributable cash flow rises, coverage improves, and the money left the building either way.

Test that split against the assets it is meant to sustain. MPLX's maintenance capital is 3.6% of adjusted EBITDA; Enterprise guides to roughly $580 million on $9,964 million, near 6%. The gap is wide, and part of it is honest, because gathering pipe and processing plants do not carry the sustaining bill of an NGL fractionation and export system. It also does not break the ratio on its own: move MPLX to Enterprise's share and distributable cash flow falls by about $170 million, leaving coverage at roughly 1.36×, which still rounds to the same headline.

What the ratio survives, the cash statement does not flatter. MPLX's own adjusted free cash flow was $1,033 million for 2025 before distributions and negative $2,991 million after them. That measure deducts every dollar of capital whatever it was labelled, so it cannot be reclassified, and a negative answer means the payout and the building programme together were funded from debt or units rather than from the year's cash. A partnership can print coverage above 1.0× for years while doing exactly that. Read the two figures together or the ratio tells you only half of what 2025 cost.

Valuation Framework

EV/EBITDA screens start against the AMZI/AMEI anchors (8.57× and 10.94× forward respectively) and the Wells Fargo 9.0× trailing median. Sponsor-linked MLPs sometimes trade below pure-play peers where the market prices conflict or concentration with the refiner parent. That discount is not automatically an opportunity. Where a sponsor can set commercial terms on both sides of a contract, a lower multiple is a price rather than a bargain.

Coverage and yield come next. At 1.4×, MPLX is the floor anchor in the coverage ratio guide. Raising the distribution from here needs faster growth in distributable cash flow, or acceptance of coverage below the 1.4× screen. Compare to Enterprise's 1.7× headroom before underwriting distribution growth.

Sponsor exposure belongs in the same workbook. Model Marathon-related throughput and contract renewals explicitly. A refiner downturn can compress gathering volumes even when tariff structures are fee-based, if minimum volume commitments are not binding on weak assets.

What to Watch in the Financials

Quarterly coverage trend. FY2025 was 1.4× but Q4 printed 1.3×, and because the higher distribution rate now applies to every quarter, 1.3× is the run rate rather than the exception. Below 1.0× the payout is being drawn from the balance sheet rather than covered. That is not the same as a cut, and a partnership can sit under 1.0× for a quarter or two on a turnaround, but it is the point at which the ratio stops reassuring anybody.

Dropdown announcements and financing. Equity-funded drops differ from debt-heavy drops for leverage and coverage paths. Watch total debt divided by LTM adjusted EBITDA against the ~4.0× supported range after each transaction.

Marathon commercial flows. Intercompany revenue and tariff disputes between sponsor and partnership appear in related-party footnotes. Changes in Marathon's refining throughput or crude slate can shift MPLX gathering volumes.

The maintenance capital line, across several years. Coverage improving in a year when maintenance capital fell as a share of EBITDA is a footnote worth opening. The DCF vs FCF guide explains why the partnership metric differs from GAAP cash flow, and why this line is the one that moves it.

Peer Context

Enterprise Products is the unsponsored baseline: 1.7× coverage, 82% fee-based gross operating margin, explicit leverage target. MPLX trades tighter coverage and higher implied yield for sponsor linkage.

Energy Transfer offers scale without a refining parent, with computed coverage near 1.80× but more structural complexity. MPLX is easier to map on sponsor-linked contracts; strategic optionality is narrower than at a diversified peer.

Kinder Morgan shows the C-corp alternative. It reports free cash flow of $2,891 million and no distributable cash flow at all, so a coverage ratio cannot be built for it from the filing; dividend cover on free cash flow is the substitute. MPLX has not made that reporting change, which is convenient for comparison and worth remembering is a choice.

Key Risks

Sponsor conflict and related-party concentration. Marathon's 63.7% ownership and control of the general partner put alignment and conflict in the same capital structure. It is the partnership's largest customer and the largest recipient of the distribution that customer's throughput helps fund, and the coverage ratio shows none of that. Dropdown pricing and commercial terms deserve scrutiny in any diligence model.

Coverage at the comfort floor. 1.4× leaves little buffer for maintenance capital surprises or volume softness, and the raised distribution puts the run rate nearer 1.3×. Below 1.0×, distributable cash flow no longer covers the payout without external funding.

Refining-cycle linkage. Gathering and logistics volumes correlate with Marathon's operational choices even when contracts are fee-based. A weak refining margin environment can still slow dropdown pace and growth capex at the sponsor.

Contract quality does not reduce to one number. Enterprise discloses that more than 82% of gross operating margin is fee-based; MPLX files no equivalent, so the two cannot be lined up. That is less of a loss than it looks, because a fee-based share tells you nothing about what stands behind the fee. Read the minimum volume commitments and the credit of whoever owes them.

Structure and tax. A unitholder receives a Schedule K-1, not a 1099. Much of the distribution is return of capital, which reduces cost basis rather than being taxed on receipt, and is recaptured on sale. Unrelated business taxable income makes an MLP awkward inside a US retirement account, and non-US holders face withholding on distributions and on sale proceeds. The MLP vs C-corp guide works through it.