Downstream refining and crack spreads

A refiner earns a margin, not a price

The first thing to say about refining, and the thing that separates a prepared candidate from an unprepared one, is that a refiner does not benefit from higher oil prices. It buys crude and sells products, so the price of crude is its input cost. What it earns is the difference between the two, and that difference has its own dynamics.

This is why the classic ranking question, "crude rises twenty percent, rank these four companies," has an ambiguous answer for the refiner and a definite answer for the other three. A producer benefits. A fee-based pipeline is roughly indifferent. A service company benefits with a lag. The refiner depends entirely on whether product prices rose as much, more, or less than crude. That ranking exercise, and where each sub-sector sits, is laid out in the energy sub-sector map. Saying "it depends on the crack spread, not the crude price" is the correct answer, and it is a better answer than picking a direction confidently.

What actually happens inside a refinery

Crude oil is not one substance. It is a mixture of hydrocarbons of different molecular weights, and a refinery separates and reconfigures them into products people actually buy: gasoline, diesel and other distillates, jet fuel, and at the heavy end residual fuel oil, asphalt, and petrochemical feedstocks.

The first stage is distillation, which separates the crude into fractions by boiling point. That alone produces a product slate determined by the crude's natural composition, and it typically contains more heavy material than the market wants. So more sophisticated refineries add conversion units that break heavy molecules into lighter ones and treatment units that remove sulfur and other contaminants.

Two crude characteristics drive everything downstream of this. Density, described as light or heavy, determines how much of the barrel naturally yields high-value light products. Sulfur content, described as sweet or sour, determines how much treatment is required to meet product specifications. Light sweet crude is easier to process and commands a premium. Heavy sour crude is cheaper and requires more capable equipment to turn into valuable products.

That is the entire logic of refinery complexity. A more complex facility, with more conversion and treatment capacity, can buy discounted heavy sour crude and still produce a high-value light product slate. A simple facility must buy expensive light sweet crude to produce a comparable slate, or accept a lower-value product mix. The complexity advantage is therefore the ability to capture the discount between crude grades, and it is worth real money whenever that discount is wide and worth much less when grades trade close together.

The crack spread and what it leaves out

The crack spread is the market's shorthand for refining margin: the difference between the value of the refined products a refinery sells and the cost of the crude it buys, expressed per barrel. The name comes from cracking, the process of breaking heavy molecules into lighter ones.

Because refineries produce multiple products, standardized proxies exist. The most commonly cited is the three-to-two-to-one spread, which assumes three barrels of crude produce two barrels of gasoline and one barrel of distillate, and computes the resulting margin. It is a convention, not a description of any real facility.

That convention is exactly why the published spread is a directional indicator rather than a company's actual margin, and being able to list what it omits is a strong answer.

Energy cost deserves a specific mention because it surprises candidates. Refineries consume large amounts of energy, often natural gas, to run their processes. So a refiner's operating cost is itself exposed to a commodity price, and a spike in gas prices compresses refining margin independent of what crude and product prices are doing.

Utilization, turnarounds, and inventory

Three operational realities shape reported results.

Utilization is the share of nameplate capacity a refinery actually runs. Because a large portion of the cost base is fixed, margin per barrel and volume both drive earnings, and a facility running well below capacity earns poorly even at a healthy spread. Utilization responds to margin, since operators run harder when spreads are wide, but it is also constrained by maintenance and by unplanned outages.

Turnarounds are the scheduled shutdowns during which units are inspected, repaired, and replaced. They happen on multi-year cycles, they are expensive, and they take capacity offline for extended periods. A quarter containing a major turnaround will look weak for reasons that have nothing to do with the market, which is why analysts adjust for turnaround schedules when comparing periods. A candidate who mentions turnarounds unprompted signals real familiarity.

Inventory accounting produces a distinctive earnings effect. A refiner holds substantial crude and product inventory, and depending on the accounting convention it uses, a period of rising prices can produce inventory holding gains that flatter reported earnings, while falling prices produce holding losses and potential write-downs. These are real in an accounting sense and misleading as a read on operating performance, so analysts often look at a margin measure that strips them out. This is the same instinct required elsewhere in the sector, where reported figures need translation before they mean anything, as with exploration accounting in energy accounting: successful efforts versus full cost.

Why the earnings swing so hard, and how to value through it

Refining margin is a spread between two volatile prices, and a spread is more volatile than either of its components. Add high operating leverage from a largely fixed cost base and the result is earnings that can swing from very strong to negative within a cycle, at the same physical assets.

This makes a spot multiple close to meaningless, and the way it fails is counterintuitive enough that interviewers ask about it. At the top of a margin cycle, trailing earnings are high, so EV to EBITDA looks low and the stock screens cheap, right before margins normalize. At the bottom, trailing earnings are depressed or negative, so the multiple looks enormous or is undefined, right when the assets may be genuinely undervalued. A naive multiple screen therefore points the wrong direction at both ends.

The standard responses are two. First, mid-cycle EV to EBITDA: estimate what the business earns at a normalized margin across a full cycle and apply a multiple to that, which pushes the real argument into what mid-cycle means, where it belongs. Second, replacement cost, valuing the facility against what it would cost to build comparable capacity, adjusted for complexity so that a more capable refinery is worth more per barrel of capacity. Free cash flow yield is a useful supplement because refining is capital intensive and turnaround spending is a recurring real cost that a pure EBITDA multiple ignores.

The same normalization problem appears in oilfield services for the same underlying reason, since service earnings swing even harder than refining margins. The full multiple-by-multiple comparison across the sector, including how services get normalized, is in how energy companies are valued.

What makes one refinery worth more than another

Four variables carry most of the difference, and naming them cleanly is a good answer to a comparison question.

Complexity, as described above, determines which crudes the facility can process profitably and therefore how much feedstock discount it can capture.

Location determines both crude access and product market access. A refinery with pipeline access to advantaged domestic crude has a structural cost advantage over one dependent on waterborne imports. On the product side, proximity to a large demand center reduces distribution costs and can support better realized prices.

Scale matters because fixed costs spread over more barrels, and large integrated sites can share utilities and infrastructure across units.

Integration with marketing or petrochemicals changes the earnings profile. A refiner that also owns retail fuel stations has a segment whose margin often moves inversely to the refining margin, since retail margins tend to widen when wholesale product prices fall. A refiner with petrochemical capacity can shift value into products with different cycles.

What interviewers ask

The ranking question described at the top is the most common, and the correct move is to flag the refiner as ambiguous and explain why.

The second is why refining trades on mid-cycle rather than spot earnings, where the answer is the inverted multiple problem.

The third is what the crack spread misses, where you want crude slate, configuration, location and logistics, energy cost, and utilization.

The fourth, which is more advanced, asks what happens to a refiner when the discount between light sweet and heavy sour crude narrows. The answer is that a complex refinery loses part of its competitive advantage, because its ability to process cheap heavy crude is worth less when heavy crude is no longer cheap, so the margin differential between complex and simple facilities compresses. Getting that right demonstrates you understand complexity as an option on a price differential rather than as a generic quality score.

Practice question

Crude oil prices rise sharply. What happens to a refiner?

It depends, and I'd say that first, because unlike a producer a refiner doesn't earn a price, it earns a margin. Crude is its input cost, so a higher crude price on its own is a headwind. What determines the outcome is whether product prices, gasoline, diesel, jet fuel, rise by as much or more, which is the crack spread. If products move up with crude, the margin is roughly intact and the higher price level is mostly neutral. If crude moves faster than products, margins compress and the refiner is worse off. There are a few second-order effects worth mentioning. If the company uses an inventory accounting convention that produces holding gains, rising prices can flatter reported earnings even when the underlying margin didn't improve, so I'd want to look at a margin measure that strips that out. And the specific refinery matters: a complex facility that runs discounted heavy sour crude behaves differently from a simple one that needs light sweet, and if the discount between those grades narrows at the same time, the complex refinery loses part of its advantage. So my answer would be that the crude price alone doesn't tell you the direction, the crack spread does, and the company-specific answer depends on configuration, crude slate, and location.

What the interviewer is listening for: Refusing to give a direction and going straight to margin versus price is the whole point of the question. Bonus credit for the inventory holding gain effect and for treating complexity as an option on the light-heavy discount rather than as a fixed quality score.