A charge can have a familiar name and still answer an unfamiliar question. A fuel surcharge sounds like a repayment of a fuel bill. In a transport contract, however, it may be a calculation attached to a reference price, a previous observation period and a specified billing base. The customer sees one additional amount. Behind it sit several decisions about what is being measured, when the measurement enters the invoice and which activity receives the charge.
That distinction matters when fuel prices move quickly. A shipper comparing an invoice with today's diesel price may be comparing different months. An analyst comparing surcharge receipts with fuel expenditure may be comparing different quantities and coverage. Neither comparison is useless. Both become misleading when the difference is assigned a meaning before the underlying definitions are examined.
In the United States, an August 2026 Reuters report put this issue in focus. It reported that Union Pacific's second-quarter fuel surcharge receipts exceeded its fuel expenditure by $91.1 million, following a first-quarter shortfall. The company described surcharges as part of the overall transportation price negotiated with customers.
The useful analytical question is not whether the word “fuel” promises an exact reimbursement. It is what the agreed formula actually links together. Understanding that link requires separating a benchmark from an expense, a billing period from an observation period, and a chargeable base from the physical consumption of diesel. The discussion below examines those distinctions; it does not reconstruct the company's accounts or assess the legality of a particular charge.
Start with the unit behind the charge
A physical expense can be written as quantity purchased multiplied by price paid, subject to whatever accounting boundary is being used. A surcharge may instead multiply a contractual rate by distance, or apply a percentage to a freight charge. These expressions do not contain the same variables. Giving both amounts a fuel-related label does not make them interchangeable.
Consider the questions that a genuinely matching calculation would have to answer. Is the expense for fuel purchased or consumed? Does the receipt cover the same movements? Is the published index the carrier's acquisition price? Are empty movements included in the consumption figure but absent from the billed base? These are questions to investigate, not claims about any particular railroad's practices.
A public total may not provide those answers. It can reveal a difference worth examining without explaining every component. That is an important middle position. Rejecting an automatic interpretation is not the same as dismissing the difference. It means identifying which conclusion the evidence supports and which conclusion still needs a reconciliation.
The reference price is a measuring device
An index provides a shared observation that both parties can identify. It need not reproduce an individual supplier's purchase price, delivery arrangement or consumption pattern. Its commercial attraction can be precisely that it is independent of those private details. The trade-off is a gap between a transparent common reference and the actual circumstances of a particular operation.
That gap should not disappear from the analysis merely because the formula is easy to calculate. A perfectly reproducible invoice can still track a different exposure from the one a reader assumes. Reproducibility answers whether the calculation follows its rules. Economic correspondence asks whether those rules move with the costs or risks the parties intended to allocate.
The observation month is not the billing month
Union Pacific's published standard carload mileage program uses a diesel benchmark and applies the resulting charge beginning in the second month after the reference month. Its separate rate-based carload program applies a percentage to referenced line-haul charges. Both pages distinguish these arrangements from intermodal surcharges; they are not universal instructions for every shipment.
The timing distinction is intuitive once two calendars are put side by side. One records when the reference price was observed. The other records when that observation affects a shipment's bill. Looking only at the latter hides information about the former. A charge can therefore rise while a current market quotation falls, without the two series being mathematically inconsistent.
This does not establish that timing explains the whole reported difference between receipts and expenses. It identifies one mechanism that can create a difference. Changes in traffic, the billing base, actual purchasing prices and the selected reporting window can matter as well. A mechanism is not a complete attribution.
A deliberately simplified timing example
Imagine an invented arrangement with one-period delay, not the two-month schedule described above. The reference level is 100 before the example begins. Over three successive periods it becomes 100, then 160, then 100 again. Suppose, solely to isolate timing, that current exposure equals the current reference level and the delayed charge equals the previous period's level. Quantity is identical in every period.
In the first period, charge and exposure are both 100. In the second, the exposure is 160 while the charge remains 100: a gap of minus 60. In the third, exposure returns to 100 while the charge rises to 160: a gap of plus 60. Across these three deliberately constructed periods, charges and exposure both total 360.
The cancellation is a property of the assumptions, not a promise about transport contracts. The model uses equal quantities, the same benchmark for the underlying exposure and the charge, a known opening reference and an observation window that includes the reversal. Change one of those conditions and cancellation need not follow. Ending the analysis after the second period would show only the shortfall.
Now change only the quantity in the third period to two units. Its charge becomes 320 and exposure becomes 200, producing a surplus of 120. Together with the earlier shortfall of 60, the three-period result is a surplus of 60. Nothing in this invented example requires a new formula. Different activity weights are enough to change the cumulative answer.
These are abstract units, not dollars from a carrier's filing and not a forecast. Their purpose is to show why a reader cannot assume that a lag always cancels within a quarter or a year. The relevant observation window and activity weights must be established rather than selected because they produce a convenient result.
The model has another limitation: it deliberately makes current exposure equal to the external reference. In a real analysis, that correspondence would need to be established. If the purchase price follows a different path from the index, another gap appears that cannot be explained by moving the observation month. The demonstration must therefore remain a demonstration rather than replace verification of the actual relationship.
A percentage and a distance charge respond to different changes
Timing is only one dimension. The base receiving the surcharge also matters. Take a fictional surcharge of 10 percent applied to a freight charge of 1,000 monetary units. The additional amount is 100. If the underlying freight charge rises to 1,200 while the percentage stays unchanged, the surcharge becomes 120. The extra 20 follows from the billing base, not from a change in the assumed fuel index.
In another fictional arrangement, charge 0.10 monetary units per distance unit across a billed distance of 1,000. The surcharge is also 100. Raising the underlying freight price alone does not change this calculation. Changing the billed distance does. Two arrangements can start with the same amount while having different sensitivities to later commercial changes.
Neither example establishes that one basis is inherently fairer. A distance measure does not by itself describe load, terrain, equipment, empty running or fuel efficiency. A percentage of a freight rate does not by itself describe gallons consumed. Choosing a base is a way of allocating exposure through a contract, not a direct observation of every physical input.
The examples also should not be combined into a supposed actual customer bill. They omit thresholds, exemptions, routing provisions, minimums and other terms that a real agreement may contain. Their controlled simplicity allows one variable to move at a time. Actual invoice checking requires the applicable documents, not substitution of this article's invented numbers.
Keep the comparison fixed before changing one variable
A useful comparison begins with the same movement, volume, service scope and reference period. Only then should one change the index, base price or distance. Changing several elements simultaneously may reproduce the final invoice but conceal why it changed. A bridge between the old and new amount is more informative than an unexplained percentage increase.
For example, a buyer could distinguish the portion associated with a changed benchmark from the portion associated with a changed base freight price. That is an analytical decomposition, not an allegation that either component is improper. Its value is practical: the parties can discuss the variable that actually moved instead of arguing over the label attached to the total.
The order of that decomposition should also be explicit. When both a rate and its base change, allocating the combined effect between separate lines depends on the comparison method. The final amount remains the same, but the explanation of individual contributions can differ. State the method in advance and use it consistently rather than presenting one convenient allocation as the only possible interpretation.
Quarterly totals answer a different question from shipment invoices
An aggregate comparison pools many movements and possibly several charging arrangements. A single shipment calculation cannot explain that pool unless its scope and weight are representative. Conversely, an aggregate surplus does not establish what happened on a particular lane or customer account. Moving between those levels requires additional information.
The distinction between receipts minus fuel expense and profit is especially important. The subtraction compares two selected amounts. It does not include every cost of providing transportation or establish the incremental earnings associated with a particular pricing decision. Calling that difference total profit would silently introduce claims the calculation does not contain.
It would be equally premature to call the whole difference a harmless calendar effect. That explanation also requires evidence. A responsible reading leaves the gap visible, examines the mechanisms that could contribute to it and avoids assigning a complete cause without a matching bridge. Both dismissive and accusatory shortcuts can outrun the data.
Reporting boundaries deserve attention as well. An opening period may carry charges based on earlier observations; a closing period may leave later adjustments outside the window. Extending the window can change the answer, but choosing a longer window does not automatically make it the correct one. The period should fit the question being asked.
Identical calendar boundaries do not guarantee identical economic coverage. One measure may include a different set of activities from the other. Establish the scope as well as the dates before comparing them. If the necessary detail is unavailable, preserve that limitation in the conclusion. Missing information should not quietly become an assumption that the two scopes match perfectly.
A review that separates definitions from disagreements
Before debating whether a result is commercially acceptable, the parties need a common description of the calculation. A short review can establish that foundation without pretending to settle a legal issue or reconstruct confidential operating costs:
- Identify the exact charging program referenced by the applicable agreement.
- Record the benchmark observation period separately from the shipment and invoice dates.
- State whether the base is a freight amount, distance or another defined quantity.
- Separate a change in the benchmark from a change in the base or shipment mix.
- Check whether compared receipts and expenses cover matching activities and periods.
- Label any remaining unexplained difference instead of assigning it automatically to profit or timing.
This sequence distinguishes arithmetic verification from commercial judgment. The first asks whether the amount follows the specified formula. The second asks whether the formula remains suitable for the relationship. Correct arithmetic does not end the second conversation, while dissatisfaction with an outcome does not establish an arithmetic error.
A purchaser may prefer predictable updates because frequent changes complicate budgeting. A carrier may prefer a reference that responds more quickly to cost movements. Those preferences can coexist with a shared interest in clear definitions. Transparency does not require identical commercial interests; it requires that each party can see which risk the arrangement leaves with them.
Internal communication matters too. A procurement manager, financial analyst and invoice reviewer may use the same word while pursuing different questions. The first needs comparable offers, the second an explanation of changing expenditure, and the third compliance with a specific term. A shared set of definitions reduces the chance that the result of one review will be mistakenly extended to the other two.
Read the mechanism before reading the headline
The broader lesson is modest but useful. A fuel-related charge can be transparent without matching every fuel expense, and a visible difference can be important without proving a complete explanation. The analytical task is to preserve those distinctions rather than compress them into a single judgment.
For an operator, the relevant question is how the chosen formula behaves when prices, traffic and base rates move differently. For a customer, it is which part of a changing invoice comes from each contractual input. For an outside reader, it is whether the available evidence supports a statement about arithmetic, exposure, aggregate recovery or profitability. Those are related questions, but they are not the same question.
A clearer discussion therefore starts with three labels: reference index, reference month and billing base. Once those are explicit, a surprising result becomes something that can be investigated. Without them, even an accurately reported number can become the starting point for an interpretation that the underlying calculation never established.