Global Partners LP GLP

Key data

The business

Global Partners is a Northeast fuel business in three parts. It owns or supplies about 1,600 gas stations and convenience stores, the biggest piece, sold under names like Alltown; it runs a fleet of storage terminals that take fuel off ships and pipelines and hand it to wholesalers; and it sells gasoline, diesel, and heating oil in bulk. The money is made two ways: the spread between what it pays for fuel and what it charges at the pump and the rack, and the higher, steadier margin on everything sold inside the store, where a candy bar earns more than a gallon. Over the last two years it has bought terminals and stations from larger oil companies pulling back, buying throughput cheaply while the sellers were motivated. The moat is real but modest: dense terminal and station coverage in one region that a newcomer cannot easily rebuild, wrapped in a partnership that must keep feeding a large cash distribution to its unit holders every quarter.

The numbers

Revenue at a fuel distributor is mostly the price of oil passing through, so read the profit, not the top line.

The five quarters show the inflection cleanly. Revenue drifted around $4.6B through 2025, then jumped to $5.3B and $6.8B as acquired terminal volume landed. Net income did more than revenue, near tripling from the mid-$20Ms to $71M. Gross profit rose 30% in Q1 and 21% in Q2 year over year, and operating income climbed with it, so the jump is operational, wider product margin and more of it, not a below-the-line item.

The annual picture is why the multiple matters. Profit fell four straight years from the $362M of 2022 to $98M in 2025, the ordinary grind of a fuel distributor as spreads normalized. Then H1 2026 alone booked $141M, beating the whole of last year by half. Either the business stepped up a level, or it is having its best margin year in four. This is the same movie refiners and distributors always play: fat spreads look permanent right up until they aren't.

Consensus takes the second view. It puts FY2026 EPS at $5.87 and FY2027 at $3.80, a planned 35% giveback, so the forward multiple reads 9x on this year and about 14x on next. On cash flow the partnership sits at 10.6x EBITDA, the 83rd percentile of its twelve-year range and above the 10.2x peer mark. What this memo believes the tape does not: the acquired terminal and store base should make the 2027 fall shallower than modeled. The one print that settles it is second-half product margin per gallon.

Management

The clearest tell is the general partner buying its own units in the open market, $1.25M across three May 2026 purchases, part of $5.3M of insider buying against $1.5M of sales over the year, net buyers by better than three to one. The Slifka family runs the partnership and the incentive rights were retired years ago, so unit holders and insiders sit on the same side of the table. Buybacks are token, about $10M last year; the cash goes to the distribution and to debt. Plan status on the insider sales is not disclosed, so treat them as routine rather than a signal. Guidance has been beaten hard two prints running, $1.86 against a $1.24 estimate last quarter.

How it fails or surprises you

Fuel margins revert to normal. Analysts already model it: FY2027 EPS $3.80 against FY2026's $5.87, a 35% drop, on the view that last year's wide per-gallon spreads fade. If station product margin per gallon reverts to its 2024 level, the $141M half-year pace roughly halves. The print that settles it is Q3 2026 station margin per gallon against Q3 2025. Everything turns on that one line.

Leverage meets a soft year. Net debt is about $2.0B, near 4.9 times EBITDA, at the high end of where this partnership runs. The 10-Q's own reserves language warns cash from operations may not be enough to hold the distribution at current levels. If margins soften while rates stay high, debt paydown and the payout compete. Watch interest coverage, 8.4x today, over the next two quarters.

The roll-up is structural, not a spike (right tail). The case for owning it is that the profit leap is bought, not borrowed from a good year: terminals and stores acquired from retreating oil majors now throw off durable throughput and steady in-store margin. If gross profit holds as gallons slowly fall, the market re-rates it from gas-station cyclical toward regional infrastructure. Nobody pays for that today. First sign: terminal and wholesale margin dollars rising into Q4 while gallons drift down.

Closing thoughts

Product margin per gallon in the second half settles it. The market has already voted: FY2027 earnings pegged 35% below FY2026 says it reads H1 2026 as a fuel-spread peak that reverts. The whole disagreement is cyclical spike versus structural step-up. An in-between print, margins easing but not collapsing, throws the decision back onto the balance sheet, and there the 4.9x leverage and the distribution warning are what you would be left holding. The fatter tail is the downside, because you are paying a top-of-range multiple on what is probably a peak year.

The bet is still that Global Partners keeps buying stations and terminals across the Northeast and earns more on each gallon and each candy bar than the year before. What breaks it is fuel margins reverting while $2.0B of debt sits at 4.9x cash flow. The one pair of numbers that tells you first: station product margin per gallon, and net-debt-to-EBITDA, in the next two quarterly reports.