Industry Guides

    What Drives Gas Station Demand and Profitability: A Feasibility and DSCR Framework

    November 6, 2026
    50 min read
    Gas station forecourt and convenience store feasibility analysis

    Most gas station loans are still underwritten on gallons, and gallons have become the least reliable number in the file. This framework ranks the six drivers of demand and profitability and follows them through to debt service coverage.

    Summary

    Most gas station loans are still underwritten on gallons, and gallons have become the least reliable number in the file. U.S. gasoline demand peaked in 2018 and has drifted down about 1 percent a year since. Fuel produced 65 percent of the average convenience store’s sales in 2025 but less than 39 percent of its gross profit. The retail fuel margin that carried the industry from 2022 through 2025 sits close to double its pre-2020 level. This analysis ranks the six drivers behind gas station demand and profitability by what the evidence actually measures, shows how each one moves the income statement, and follows them through to [debt service coverage](/insights/dscr-feasibility-studies-complete-guide), where a site that shows 1.26x on paper can fall below the [SBA floor](/services/sba-feasibility-study) on a 2.7-cent price response from one new competitor.

    A gas station is three businesses sharing one parcel. The forecourt sells a commodity at a margin set by the street. The store sells packaged goods at a margin set by the operator. The kitchen, where there is one, sells food at margins closer to a restaurant’s than to a fuel retailer’s. Feasibility studies still tend to treat the forecourt as the business and the store as an amenity. Twenty years ago that was a reasonable shorthand. It is not now.

    The 2025 numbers make the case. Industry fuel sales fell 5.4 percent to $476.3 billion while in-store sales reached a record $341.2 billion, the 23rd consecutive annual increase (11). Fuel was 65.0 percent of the average store’s sales and 38.8 percent of its gross profit; merchandise and foodservice together produced 57.4 percent (11). Gallons per store rose only 0.5 percent. Over the same period the retail gasoline margin averaged 39.7 cents a gallon in both 2024 and 2025, against roughly 22 to 24 cents a gallon between 2014 and 2019 (23). A study that projects flat gallons at today’s margin is underwriting two favorable conditions at once, and neither one is structural.

    For a lender the question has moved. It is no longer how many cars pass the site. It is how much of the debt service the store covers, how much depends on a fuel margin that has never been this high for this long, and what happens to both when gallons erode at the rate every official forecast now assumes. The sections below take the drivers one at a time, then rebuild them into a profit and loss statement, a feasibility forecast and a coverage test. Every figure is dated and sourced, and where the public record has a gap we say so rather than fill it.

    Three businesses on one parcel

    Start with where the money is made, because it decides which drivers matter.

    The profit pool has moved inside. In NACS’s 2025 data, foodservice was 28.5 percent of in-store sales and 38.9 percent of in-store gross profit; in 2005 it was 11.9 percent of in-store sales (11). At the operator level the mix depends almost entirely on format. Casey’s, a rural Midwestern chain with a pizza kitchen in nearly every store, earned 34.6 percent of its fiscal 2026 gross profit from fuel, 37.8 percent from grocery and general merchandise and 24.1 percent from prepared food and dispensed beverage, by our arithmetic on its reported category totals (16). Murphy USA, a small-box, low-price format built beside Walmart stores, reported $1,488.7 million of fuel contribution against $869.0 million of merchandise contribution in 2025 (15). Both are healthy businesses. They are not the same business, and a study that applies the national average mix to either one will misstate its profit by a wide margin.

    Gallons are a traffic input; margin is the profit input. Fuel gross profit is gallons times cents per gallon, and the second number moves more than the first. NACS puts a retailer’s cost of selling fuel at roughly 20 to 25 cents a gallon, about 8 cents of it card fees, which leaves 10 to 15 cents of net margin on a 39.7-cent gross (11, 12). At that spread, a 5-cent change in the street margin is a third to a half of the fuel line’s net contribution.

    The store has its own traffic problem. Total transactions per store fell 2.7 percent in 2025 to 1,484 a day, and inside transactions fell 1.6 percent per store per month (11). NACS Research reported in-store transactions down 4.5 percent in the first half of 2026, a decline its managing director called the worst he had seen with both sides of the business off at once (41). The average inside transaction swung from a one-cent profit in 2024 to a seven-cent loss in 2025 after fully allocated expenses (11). Inside sales are still growing, but the growth comes from basket size and food, not from more visits.

    The practical rule for underwriting follows directly. A gas station forecast has at least three revenue lines, each with its own drivers: fuel gallons times net margin; merchandise sales times merchandise margin; and foodservice sales times food margin, constrained by kitchen and labor capacity. Other income, such as lottery commissions, car wash revenue and rent from a quick-service tenant, sits beside them. The table below shows where each of the six drivers acts first and how well its effect has been measured. The ranking of evidence quality is roughly the reverse of the order in which most studies present the drivers.

    DriverWhat it moves firstHow well the effect is measured
    Fuel demand fundamentalsGallons per capita over the loan termStrong: EIA, FHWA and EPA series
    Price, brand, loyalty and hypermarketsGallons at the site and the fuel marginStrong on price; no public study of volume lost to club entry
    Inside sales and foodserviceIn-store gross profitStrong at industry and operator level; no causal study of food-driven trips
    Supply, consolidation and regulationMargin, entitlement and incumbent defensibilityStrong counts and price effects; no public volume-cannibalization study
    EV transitionLong-run gallons and residual valueStrong on adoption; little evidence on site-level impact
    Traffic and site captureGallons at a given siteWeak: rules of thumb, no audited capture standard

    Driver 1: national fuel demand has peaked

    The national demand curve is the backdrop every site forecast sits on, and it now slopes down.

    The peak is established, not speculative. EIA puts the record for U.S. gasoline consumption at about 392 million gallons a day in 2018, roughly 9.3 million barrels a day (1). Consumption in 2025 averaged 8.9 million barrels a day, 1 percent below 2024 and 4 percent below 2019, about 136.5 billion gallons for the year (1, 2). EIA’s October 2026 Short-Term Energy Outlook holds motor gasoline at 8.75 million barrels a day in both 2026 and 2027 (3). Distillate, which includes on-road diesel, is roughly flat at 3.84 to 3.91 million barrels a day; it follows freight and industrial output, not households, which is why truck-stop and passenger-car sites need different volume trajectories (3).

    Driving is up; gallons are down. Vehicle miles traveled reached a record of about 3.32 trillion in 2025 (5). EIA attributes the gap to fleet fuel economy, which rose 1.9 percent in 2025 against 1.2 percent more driving, and it assumes miles per gallon will keep improving about 1 percent a year in its forecast (2). The mechanism is fleet turnover. New vehicles averaged a record 27.2 miles per gallon in real-world terms in model year 2024, with a preliminary 28.1 for model year 2025, while the on-road light-duty fleet averaged 23.4 in 2024 (6, 5). Every year that newer vehicles replace older ones, gallons per vehicle fall. Hybrids, which still buy gasoline but much less of it per mile, reached a record 16.3 percent of new sales in the second quarter of 2026 (26). Vehicles stay on the road 18 to 28 years, so this effect runs for a decade or more regardless of what happens in Washington (4).

    Policy moved sharply in 2026, and it slows the decline without reversing it. EPA rescinded the 2009 endangerment finding and all vehicle greenhouse gas standards in a rule published February 18, 2026 (8). NHTSA’s SAFE Vehicles Rule III, published September 30, 2026, resets fuel economy standards to roughly 34.9 miles per gallon for model year 2031, against about 49.3 under the 2024 rule as NHTSA now estimates it (7). Both face litigation. EIA does not expect the rollback to change fleet fuel economy much in the near term because vehicle product cycles run five to seven years (2).

    The official forecast cases all point down. In EIA’s Annual Energy Outlook 2026, released April 8, 2026, gasoline consumption in 2050 is lower than in 2025 in all 11 cases (4). The case EIA now calls the Counterfactual Baseline, formerly the Reference case, assumes vehicle standards that were repealed two months before publication, and EIA itself says it should not be regarded as the most likely outcome (4). The Alternative Transportation case, which drops those standards, carries 2.3 million barrels a day more gasoline in 2050 and is the closer match to current law (4). A secondary computation from the AEO data file puts the Baseline’s decline near 2.5 percent a year and the Alternative Transportation case near 1 percent a year; those rates are our arithmetic on secondary values and should be treated as approximate (4).

    Remote work redistributed trips rather than removing them. The share of workers who usually work from home rose from 5.7 percent in 2019 to a peak of 17.9 percent in 2021 and was 13.3 percent in 2024, and the share driving alone to work fell from 76 to 69 percent (9). Total driving still set records. Trips moved toward midday, errands and suburban neighborhoods. NACS survey data show 33 percent of drivers buy fuel between 10 a.m. and 3 p.m., more than in either commute window (14). For site selection that favors neighborhood-serving locations with daytime residential density over sites whose thesis is a downtown-bound commuter stream.

    Regional spread is large and structural. Per-capita gasoline consumption in 2018 ran from about 552 gallons in Wyoming and North Dakota to about 284 in New York and 164 in Washington, D.C., against a national 422 (10). EIA attributes the spread to density, trip length and transit access; price differences have only a limited effect (10). The same pattern shows up in store data. Southeastern stores sold 186,536 gallons a month in 2025, 18.4 percent above the national average, while Northeastern fuel sales per store fell 9.3 percent against 5.4 percent nationally (13).

    Two cautions apply to 2026 itself. Retail gasoline averaged $4.35 and diesel $6.29 a gallon in September 2026 after a Middle East supply shock, and EIA expected second-half consumption to run below its five-year low in some months (3). A 2026 base year is cyclically depressed in volume and inflated in price. And EIA suspended its on-highway diesel sales series after February 2022, so there is no current federal measure of on-road diesel volume; truck-stop studies have to rely on FHWA truck fuel data and operator evidence (1, 5).

    The underwriting consequence is simple and still frequently missed. A pro forma that holds gallons flat after stabilization assumes per-capita demand stops eroding, which contradicts every official source. A defensible base case erodes per-capita gallons about 1 percent a year, in line with EIA’s fuel economy assumption and the policy-current AEO case; a defensible downside uses about 2.5 percent, the Baseline trajectory that would return if courts or a future administration restored the standards. Both should be applied from a normalized base year, and both should be carried through to coverage in year ten, not only year two.

    Driver 2: traffic, site characteristics and fuel capture

    Traffic counts lead almost every gas station study, and their statistical support is the thinnest on this list. Only two kinds of source give audited anchors for site volume: operator filings and NACS operating data. Nearly everything else is a rule of thumb.

    The thresholds are screens, and they depend on format. Traffic-engineering guidance puts the traditional c-store range at roughly 5,000 to 15,000 vehicles a day on the adjacent road; the ceiling comes from what access permits allow, not from demand, because above about 15,000 most agencies restrict a site to three-quarter or right-in, right-out access (19). The Appraisal Institute literature gives 2,000 to 15,000 on connector roads (19). Large-format operators publish much higher floors: Sheetz asks for average daily traffic at or above 18,000, and Wawa and Refuel ask for at least 25,000 vehicles a day with access to the site (18). No national truck-stop chain publishes a traffic minimum. The traditional ceiling sits below the large-format floor, which is not a contradiction; the formats need different trade areas and different access. A site that clears a threshold has only earned the right to be modeled.

    Capture rate is the weakest number in most models. There is no published standard. The figures in circulation, including ranges that practitioners like us have quoted for stabilized branded sites, run from about 0.5 to 3 percent of adjacent traffic, and some marketing material reaches 10 to 15 percent. The figures differ up to tenfold because they measure different things: fuel transactions, all store visits or pass-by trips in the numerator; one-way or two-way counts, a single road or a combined intersection in the denominator. The only published regression, in The Appraisal Journal in 2012, fits capture as an inverse power function of traffic, with an R-squared of 53 percent (19). On that curve, a site on a 10,000-vehicle road captures about 3.1 percent of passing traffic, roughly 310 customer visits a day; at 50,000 vehicles capture falls to about 0.95 percent and visits rise only to about 475, by our arithmetic on the published equation (19). Five times the traffic buys about 1.5 times the customers.

    The traffic study and the feasibility study have to agree. ITE’s average afternoon pass-by share for a convenience store with fuel is 56 percent, within an observed range of 46 to 72 percent, and research for NCDOT places pass-by at fueling centers between 45 and 80 percent (20). At NACS’s 1,484 transactions a day, a 56 percent pass-by share implies about 830 diverted visits, or 4.2 percent of a 20,000-vehicle road, for every purpose combined (11, 20). NACS’s 322 fueling transactions a day would be 1.6 percent of the same road (12). A capture assumption that implies more pass-by traffic than the traffic impact study allows is an error in one of the two documents, and a credit officer who has read both will find it.

    Gallons per transaction is the one anchored number. NACS reports 8.4 gallons per fueling transaction and 322 fueling transactions per store per day in 2025, about 2,700 gallons a day (12). Many consultant models and sponsor pro formas use 10 or 11 gallons per fill, which overstates volume by a fifth or more on the same transaction count. No authoritative average exists for Class 8 diesel fills; tank capacities and anecdotes are not a benchmark, so truck-stop diesel should be built from truck counts, fleet-card exposure and pricing position.

    Site characteristics have direction but almost no published effect size. Price beats location in every survey: 72 percent of drivers name price as the main factor in choosing where to buy fuel, against 16 percent for location and 12 percent for brand (14). Wawa requires a corner at a signalized intersection with visibility from each approach at the posted speed (18). The 15 to 30 percent capture loss practitioners apply to right-in, right-out access is a working rule, not a published standard. The best-known measured case, a Texas study of five stations after a raised median was installed, recorded gross sales declines of 17.6 percent during construction and 40.4 percent afterward, from a tiny and dated sample (20). Going-home side, canopy visibility, ramp distance and truck share have no reliable published uplift. A study that applies one is using an assumption and should say so.

    Volume benchmarks span two orders of magnitude. Murphy USA averaged 235,751 gallons per store per month in 2025 (15). Casey’s averaged 1.217 million gallons per mature store in fiscal 2026, about 101,000 a month (16). An OPIS analyst put the busiest Costco fuel centers at up to 1 million gallons a month against about 80,000 at a typical station (23). NACS’s own national averages conflict by basis: 139,936 and 141,813 gallons per store per month for 2023 in two NACS publications, and both “3,000 to 4,000 gallons a day” and “about 2,500 gallons a day” on its fact pages (13, 12). They should not be blended. Murphy’s 2022 quartile data show how steeply volume falls below the top: about 229,000 gallons per store-month in the first quartile, 136,000 in the second and 120,000 in the third (15). An independent operator belongs in the lower quartiles until its comparables prove otherwise.

    Ramp is the least disclosed variable and the one that sinks first-year coverage. Murphy’s disclosed new-store data imply about 91 percent of run-rate fuel volume in the first 12 months, full ramp after about three years, and merchandise at only about 60 percent of run rate in year one (15). Casey’s describes a positive return in year one, a double-digit after-tax return by year two and mid-teens by years three to four (16). Representative independent-site models in practitioner literature assume closer to two-thirds of stabilized volume in year one. Murphy’s figure reflects an everyday-low-price format with an established brand beside a Walmart. For an independent, we underwrite year-one fuel at two-thirds to three-quarters of stabilized volume and year-one inside sales lower still, unless the sponsor has its own documented ramp history.

    Driver 3: inside sales and foodservice, the profit engine

    If gallons are the traffic input, the store is where the debt gets paid. It is also the line most studies forecast with the least discipline.

    Industry growth is slow, and the dispersion between operators is wide. Industry in-store sales rose from $213.5 billion in 2014 to $341.2 billion in 2025 while the store count stayed near 152,000, so per-store productivity did the work (11). The 2025 increase of 1.7 percent was the smallest since 2020 and below inflation (11). Chain results show how little the average says about a given site. Casey’s inside same-store sales rose 4.2 percent in fiscal 2026, with prepared food up 5.2 percent (16). ARKO, a tobacco-heavy acquirer with a thin food program, reported same-store merchandise sales down 4.1 percent in 2025, and 2.7 percent excluding cigarettes (36). The operators with scaled foodservice are growing; the ones without it are shrinking. A site’s inside-sales trajectory should follow its operator model, not the NACS headline.

    Foodservice is the margin lever. Prepared food earned a 56.41 percent gross margin across NACS’s 2024 sample, and Casey’s reported 58.6 percent on prepared food and dispensed beverage in fiscal 2026 against 35.8 percent on grocery and general merchandise (11, 16). Foodservice sales averaged about $72,000 per store per month nationally in 2025 (13). The top tenth of operators by store profit had nearly three times the foodservice sales and four times the foodservice gross profit of the bottom tenth in 2024 (11). That is a correlation with profitability, not proof of causation, but it is a strong one. The leakage runs the other way too: NACS finds that a third of convenience shoppers plan to visit a quick-service restaurant within 30 minutes, a pool it sizes at more than $100 billion a year (13).

    What is not measured is whether food creates trips that fuel would not. No public industry study isolates food-driven visits from fuel-driven ones. The closest evidence is regional and operator-specific. In NACS’s Central region in 2025, inside transactions rose 13.6 percent while pump transactions fell 12.6 percent and foodservice sales grew 4.4 percent (13). Casey’s reports that guests who order wings on their own increase their prepared-food order frequency by 30 percent (16). Fueling transactions are only 28 percent of the average store’s total transactions, which is consistent with substantial non-fuel trip generation without proving where it comes from (11). A study that credits a new kitchen with traffic independent of the pumps should support it with the operator’s own comparables.

    Conversion and basket are survey numbers, not transaction data. NACS consumer surveys report that 62 percent of drivers went inside the store the last time they bought fuel in early 2026 and 64 percent in 2025, a record (12). Going inside includes paying at the register and using the restroom, so these figures overstate purchase conversion. The national average inside basket was $8.05 in 2024, and 2025 regional baskets ranged from $6.81 in the South Central region to $8.84 in the Midwest (13). No transaction-matched national conversion rate is public.

    Bigger boxes sell more because they are better stores, not because they are bigger. Yesway’s filing for its 2026 public offering compares its new 5,800-square-foot stores with its 3,400-square-foot legacy stores: inside sales of $3.1 million against $1.7 million a year, gallons of 3.1 million against 1.0 million, and store contribution of $1.4 million against $0.4 million (17). Inside sales per square foot barely moved, about $534 against $500 by our arithmetic. The 3.2-times gain in contribution came from a larger fuel offer generating traffic, a higher-margin food mix and operating leverage. NACS’s quartile data point the same way: top-quartile stores in 2019 were about 60 percent larger and roughly twice as productive per square foot (11). Both outcomes reflect operator quality and food. A 5,000-square-foot store with a traditional merchandise mix risks lower sales per square foot, not higher.

    The ratios practitioners use do not travel between formats. Inside sales per gallon, computed from operator filings, run from about $0.88 at Murphy USA to $1.00 at Yesway’s new stores and $1.70 at Yesway’s legacy stores (15, 17). That is a twofold spread, and a ratio borrowed from the wrong format can misstate inside sales by 50 percent or more. Per-square-foot benchmarks have their own trap: the NACS quartile figures that circulate in trade press, about $37 to $75 per square foot, are consistent with a monthly basis and are often misread as annual (11). No authoritative public inside-sales-per-gallon or per-square-foot benchmark exists for 2024 to 2026; the current figures sit in NACS’s paid report.

    Driver 4: price position, hypermarkets and the fuel margin

    Price is the lever that moves fuel demand between sites, and it is the one feasibility studies parameterize least.

    Station-level demand is highly elastic. The cleanest U.S. evidence is a field experiment run on 54 company-operated major-brand stations in California in 1999, in which regular-grade prices were moved two cents up or down and held for a week (21). Own-price elasticity came in at about −3.3 in the fixed-effects model and −5.6 in the preferred instrumented model, and it rose with competitive density: −2.4 where a station had fewer than 18 rivals within two miles and −8.4 where it had 27 or more (21). A more recent study of search behavior finds a typical own-price elasticity near −8 (21). Translated to today’s prices, those elasticities imply roughly 1 to 2 percent of a site’s volume per cent of sustained price difference at the central estimates, and about 0.8 to 2.7 percent across the density range. That translation is our arithmetic, it assumes 1999 elasticities still hold, and the authors themselves note that a one-week response probably understates the long-run one. No public 2020 to 2026 source states a measured volume change per cent of price gap.

    Drivers say the same thing. In NACS’s May 2025 survey, 72 percent named price as the deciding factor and 69 percent said they would drive five minutes out of their way to save five cents a gallon (14). The behavior is local. An event study of station-level California prices from 2014 to 2018 finds that a new entrant within one mile cuts incumbent prices by an immediate and persistent 2.7 cents a gallon, about 7 percent of the retail markup; the effect is concentrated within a quarter mile, fades to irrelevance beyond about seven miles, and is small where incumbents already face more than four competitors within a mile (22). A hypermarket entrant cuts nearby prices by about 5 cents (22). Exit has a precisely estimated null effect on prices except where only one or two competitors remained (22).

    Warehouse clubs are the structural share-takers. On OPIS’s visit-based measure, Costco’s share of U.S. gasoline visits rose from 2.67 percent in 2019 to 3.78 percent in 2025, and Sam’s Club’s rose by about 100 basis points to 2.72 percent; over the same period Costco widened its discount to local prices from about 25 to 33 cents a gallon and Sam’s Club from 15 to 24 cents (23). Costco operated 747 fuel stations at the end of fiscal 2025, with gasoline about 10 percent of total net sales (24). BJ’s reported comparable gallons up 10.5 percent in its second quarter of fiscal 2026 against an industry decline of about 5 percent (24). Refiner brands fell from 45.86 to 39.85 percent of visits even as their outlet count grew (23). No current public figure gives the hypermarket share of U.S. gallons, and no public U.S. study measures gallons lost at incumbents after a club opens. The price evidence and a 20 to 33 cent gap imply large diversion within the club’s draw area, but that is inference, not measurement.

    Brands have lost demand power. Unbranded outlets have drawn more visits than branded ones since June 2019, and in early 2024 branded outlets held about 61.5 percent of locations but under 43 percent of visits (23). Branded stations priced about 3 cents above the street average in August 2024 and unbranded about 5 cents below (23). Only 12 percent of drivers name brand as their primary reason for choosing a station (14). A major-oil flag still brings supply priority in a shortage, image and dispenser capital, and a loyalty tie-in. It should be underwritten as a capital and supply-security benefit, not as a volume uplift, unless the operator can show a like-for-like comparison.

    Loyalty programs are broad and shallow at the pump. Across more than 40,000 sites, 24 percent of fuel transactions in a June 2026 week presented a loyalty card, and a third of those redeemed a per-gallon discount averaging 12.8 cents (25). Large programs mainly drive visit frequency and inside spending: Casey’s Rewards had about 10.5 million members at fiscal year-end 2026, and Murphy’s Drive Rewards members averaged five trips a month in 2023 (16, 15). The only public gallons datapoint is descriptive. Cumberland Farms’ 2026 registration statement reports that loyalty members pumped about 5 percent more per fill than non-members (25). Treat loyalty as an inside-sales and frequency lever unless the operator provides member and non-member gallon data.

    Margins replaced volume as the profit engine, and that is the biggest single risk in the file. OPIS’s U.S. gasoline gross margin averaged 29.9 cents a gallon in 2021, 42.9 cents in 2022, 39.4 in 2023 and 39.7 in both 2024 and 2025, against roughly 22 to 24 cents between 2014 and 2019 (23). The industry calls it “40 cents is the new 20 cents.” The reasons are real: falling same-store gallons, store expenses up 23.3 percent since 2021, which raises the margin a station needs to break even, and scale operators that can absorb lost gallons by holding price (23). Operator figures move with them: Casey’s earned 42.6 cents a gallon excluding card fees in fiscal 2026, and Murphy’s total fuel contribution was 30.7 cents in 2025 (16, 15). But a higher margin also widens the price umbrella under which clubs and new entrants take share. Nothing guarantees that 2024 and 2025 are the new floor.

    The underwriting consequence is that a gallons forecast which does not model the subject’s posted price against its three nearest competitors and the nearest club or hypermarket is mis-specified, and a forecast that capitalizes a 40-cent margin as permanent is optimistic by construction. A defensible base case sits between the pre-2020 range and the 2024 to 2025 level, and a defensible downside tests the pre-2020 range directly.

    Driver 5: the EV transition, slower and more local than the headlines

    The EV question appears in nearly every gas station credit memo, usually as a paragraph of reassurance or a paragraph of alarm. The evidence supports neither. It supports a slow, policy-dependent erosion that matters most for residual value and most in a handful of markets.

    Adoption has reset, not reversed. Electric vehicles were about 6.4 percent of new-vehicle sales in the third quarter of 2026, down from a peak of about 10.6 percent in the third quarter of 2025, when buyers pulled purchases forward before the federal credit expired (26). Year-to-date sales through September 2026 were about 30 percent below the prior year (26). EVs were about 2.1 percent of the 299.9 million light vehicles on the road at mid-2026 (27). Hybrids, at a record 16.3 percent of new sales, matter as much for per-site gallons as battery EVs do, because they stay at the pump and buy less (26). BloombergNEF cut its 2030 U.S. plug-in sales forecast to about 17 percent, from 47.5 percent two years earlier, and named the withdrawal of federal support as the biggest factor (28).

    The federal supports are largely gone. The new and used clean-vehicle credits ended for vehicles acquired after September 30, 2025, and the credit for charging equipment ended for property placed in service after June 30, 2026 (29). Congress nullified California’s Advanced Clean Cars II waiver in June 2025; California’s challenge is pending, and a September 2026 injunction left the state’s older waivers in place (29). NEVI charging funds were partly transferred to other highway programs in the fiscal 2026 appropriations act, and at least 42 states now charge a special EV registration fee (29). Each of these could reverse through litigation or a future Congress, which is why the scenario structure in the fuel-demand section carries a downside case.

    Efficiency, not EVs, is the dominant driver of lost gallons through the mid-2030s. EIA attributes 2025’s gasoline decline to internal-combustion efficiency and hybrid sales (2). Even in AEO2026’s high-policy cases, it takes decades for a 50 percent EV sales share to become a comparable share of the fleet, because vehicles last 18 to 28 years (4). For a loan underwritten today, national EV displacement is the smaller factor in the first ten years of a 25-year term and the larger factor in the last ten.

    The risk is concentrated in a few markets. California had about 1.84 million battery EVs at the end of 2025, about 4.8 percent of its light-duty fleet; Washington, Colorado, New Jersey and Oregon followed at 2.5 to 3.4 percent, against a national 1.9 percent (29). EV share of new registrations in the third quarter of 2025, the pre-expiration peak, reached 31.5 percent in California, 29.4 percent in Colorado and 24.2 percent in Washington (29). DC fast-charger utilization is highest in Hawaii, Washington, D.C. and California (30). The rural Midwest, Southeast and South Central regions carry the least EV exposure, but they are also the most fuel-dependent for gross profit: fuel was 70.5 percent of South Central store sales and 41.4 percent of gross profit in 2025, against 64.6 and 38.5 percent nationally (13). Lower exposure, higher sensitivity.

    Chargers are an amenity, not a gallon replacement. Across about 81,000 U.S. fast-charging ports, average utilization was 15.9 percent in the third quarter of 2026 at an average price of $0.54 per kilowatt-hour (30). Casey’s had 282 chargers at 64 of its roughly 2,944 stores at fiscal year-end 2026, and its chief executive earlier described an average of about 330 gasoline transactions a day against 13 charging sessions where chargers were installed (16). No operator filing, NACS study or academic paper quantifies an inside-sales lift from charging. Revenue per charger, capital cost per port and demand charges are not disclosed by any convenience operator in the public record. A study should not underwrite charging revenue or a dwell-time basket uplift without session data from the site or a direct comparable.

    Lenders and rating agencies have no EV-specific method, and markets are not visibly pricing the risk. SBA, USDA and the Appraisal Institute address it only indirectly, through special-purpose rules, going-concern appraisal and remaining economic life (38, 39). Morningstar DBRS expects modest, slow fuel-volume declines in 2026 and a gradual long-term EV-driven decline, and flags continued pressure on single-store operators and small chains (32). Fuel-selling convenience store net leases listed at an average 5.56 percent asking cap rate against 7.63 percent for non-fuel stores in late 2025, and long-term 7-Eleven and Wawa leases asked 4.8 to 5.2 percent in the third quarter of 2026 against 6.92 percent for single-tenant net lease overall (31). The net-lease market is pricing tenant credit and lease term, not terminal fuel demand. The EV risk shows up where that market cannot see it: in the reversion value of short-term or non-credit assets, and in owner-operated going-concern loans, which is exactly what most SBA and USDA gas station loans are.

    The underwriting consequence is to model gallons as efficiency first and EV second, layer a market-specific EV overlay from state registration data in the high-adoption states, and tie the appraiser’s remaining-economic-life conclusion to that overlay. In a high-adoption metro, a 25-year amortization should be justified explicitly. In a low-adoption rural market, the file should say why the life is not shortened.

    Driver 6: supply, consolidation and local regulation

    The U.S. fuel and convenience network is not shrinking in any meaningful sense. It is being re-sorted, and the re-sorting decides how much of a trade area’s demand a given site can keep.

    Fewer stores, more of them selling fuel, more of them owned by chains. The NACS and NIQ count stood at 151,975 stores at the end of 2025, down 280 in the year and about 2 percent below its peak of 154,958; fuel-selling stores rose 768 to 122,620, an eight-year high (33). Net losses are concentrated in high-cost Northeastern states, with New York down 143 stores in 2025, while Texas added 88 and Georgia 39 (33). Ownership is still fragmented: companies with ten or fewer stores own 63 percent of the count, and chains with 500 or more own 22.2 percent (33). That long tail is the acquisition pool. Since 2020 the large deals include 7-Eleven’s $21 billion purchase of about 3,800 Speedway stores, BP’s $1.3 billion acquisition of TravelCenters of America at about six times trailing EBITDA, Casey’s $1.145 billion purchase of 198 CEFCO stores at about eleven times pro forma EBITDA, and Sunoco’s roughly $9.1 billion acquisition of Parkland (34). Couche-Tard withdrew its roughly $47 billion offer for Seven & i in July 2025 (34).

    Large-format entry is measured on price and unmeasured on volume. No public study, OPIS analysis or peer-reviewed paper quantifies what a Buc-ee’s, Wawa, Sheetz or QuikTrip opening does to incumbent gallons, inside sales or closure rates. The causal evidence is on price: the 2.7-cent incumbent price cut within one mile of a new entrant described in the previous section, with stronger effects from hypermarkets and none from exits (22). At fuel margins of 30 to 40 cents, that is about 7 to 9 percent of fuel gross margin per entrant. Studies from Mexico and Germany point the same direction, with effects fading within a few minutes’ drive (22). Volume diversion from a mega-format opening has to be modeled through fair share and labeled as an assumption, because nobody has published the measurement.

    Regulation now constrains new supply in specific places. Petaluma, California approved a permanent ban on new gas stations in 2021 that also bars adding pumps at existing stations; Santa Rosa, Windsor and unincorporated Sonoma County followed in 2022 and 2023, the county being the first in the country (35). Elsewhere the tool is spacing. Kansas City adopted a moratorium in February 2024 while it drafted a 1,500-foot separation rule; Rome, Georgia adopted 1,000-foot spacing and school, residential and floodplain setbacks in January 2026 after counting one store per 907 residents against a national figure near one per 2,200; and Kokomo, Indiana requires 5,500 feet from existing stations (35). A Denver ordinance reported effective in February 2025 and a pending Chicago ordinance would add quarter-mile and 1,000-foot rules; both should be verified against the municipal record before a study relies on them (35). The effect is double-edged. These rules lower the entitlement probability for a new site and give scarcity value to compliant existing sites, though Petaluma’s version also caps their upside.

    Tanks are a supply constraint and a credit risk. EPA’s 2015 rewrite of the underground storage tank rules added secondary containment and interstitial monitoring for new and replaced systems, under-dispenser containment, operator training and periodic sump and spill-bucket testing (37). EPA counted about 533,000 active petroleum tanks at about 190,000 facilities in September 2025, with 4,307 confirmed releases in fiscal 2025 and a cleanup backlog of 53,777 (37). California’s deadline to remove single-walled tanks passed on December 31, 2025; the state reports that more than 99 percent were closed and estimates that over 30 percent of the remaining single-walled systems have had or will have a release (37). In California, a compliance date removed supply faster than any market force could have.

    SBA loan performance depends on the denominator. Gas stations are among the largest SBA 7(a) categories on record, roughly $17.7 billion across about 21,000 loans since 1995 in third-party tabulations of SBA loan-level data (40). The same data produce very different risk headlines. The lifetime dollar-weighted charge-off rate is about 3.5 percent; the share of resolved loans that charged off is about 15 percent; and the fiscal 2010 to 2016 7(a) cohort for NAICS 447110 charged off 3.79 percent by count against 6.50 percent for all industries (40). Mature 504 vintages from 2010 to 2016 charged off about 3.3 percent against 1.8 percent program-wide (40). Post-2020 vintages look excellent because they have not seasoned. The pattern is high frequency and low severity: the real estate and tanks recover most of the balance, which makes collateral quality, environmental condition and appraisal discipline as important as the cash flow forecast. Every default figure in a credit memo should state its denominator, program and window.

    What drives profitability: the P&L line by line

    The six drivers act on different lines of the income statement, and profitability is the net of all of them. The fastest way to see which ones matter for a particular site is to build its stabilized P&L and ask what each line depends on.

    Gross profit has three sources and they behave differently. Fuel gross profit is gallons times the street margin; it is the most volatile line, because the margin is set by competitors and the wholesale market, not by the operator. Merchandise gross profit is steadier, with blended margins around 33 to 36 percent at the large chains, and tobacco declines are its main structural drag (16, 36). Foodservice gross profit carries margins in the mid-to-high 50s but needs a kitchen, trained labor and a program customers come back for (11, 16). Other income, such as lottery and money-order commissions, ATM fees, a car wash or rent from a quick-service tenant, is small but rarely zero.

    Operating expenses are mostly fixed, and two of them scale with fuel. Labor is the largest controllable cost, and store expenses across the industry are up 23.3 percent since 2021 (23). Card fees scale directly with fuel price and gallons: NACS puts them near 8 cents a gallon, part of a total cost to sell fuel of 20 to 25 cents a gallon (11, 12). When pump prices spike, card fees rise with them while the cents-per-gallon margin often compresses, which is why a price shock can hurt a station even when gallons hold. Environmental compliance is a real line: tank and line testing, release detection, operator training and financial-responsibility coverage under the federal tank rules (37). Utilities, repairs to dispensers, canopy and tanks, insurance and property taxes round out the base. Branded sites may carry brand and image fees; leased sites carry rent.

    Appraisers value the result two ways, and the two must not be confused. Going-concern value is commonly expressed as a multiple of gross profit or as net operating income over an overall rate; one appraisal firm’s presentation to the Society of Chief Appraisers cited California gross profit multipliers of 3.0 to 5.0 and operating expenses running 50 to 70 percent of gross profit (42). Platform transactions price the business: BP paid about six times trailing EBITDA for TravelCenters, and Casey’s about eleven times pro forma EBITDA for CEFCO (34). Net-lease markets price the real estate under a credit tenant at the cap rates cited in the previous section (31). Capitalizing an owner-operator’s EBITDA at a net-lease cap rate values a business at a real estate rate. SBA requires a going-concern appraisal that allocates value among land, improvements, equipment and intangibles for exactly this reason (38).

    The table below is a stabilized P&L for a representative independent fuel and convenience site, built from the benchmarks in this article. It is illustrative, not a market average, and we use it for the coverage tests in the DSCR section.

    LineWhat drives itAssumptionAnnual amount
    Fuel gross profitGallons, price position, street margin1.5 million gallons at 34¢ gross$510,000
    Merchandise gross profitStore traffic, mix, tobacco trend$1.28 million sales at 32%$409,600
    Foodservice gross profitProgram quality, kitchen capacity$320,000 sales at 55%$176,000
    Other incomeLottery, ATM, ancillarySite-specific$45,000
    Total gross profit$1,140,600
    LaborHours, wage rates, food programFixed at stabilization($360,000)
    Card feesFuel price, gallons, inside sales8.4¢ a gallon plus 2% of inside sales($158,000)
    Utilities, repairs, insurance, taxesSite, age of equipment, jurisdictionFixed($162,000)
    Other operating, including tank complianceRegulation, systems, suppliesFixed($55,000)
    EBITDA$405,600

    Three sensitivities fall out of the table. Each cent of fuel margin is worth $15,000 a year, about 3.7 percent of EBITDA. Each 1 percent of gallons is worth about $3,800 of fuel contribution after card fees, before any knock-on effect on store traffic. Each point of blended inside margin is worth $16,000. Operating expenses are 64 percent of gross profit, inside the 50 to 70 percent range appraisers cite (42), and fuel produces about 45 percent of gross profit, above the national 38.8 percent, which is typical of a fuel-heavy independent with a modest food program (11). On these numbers the street margin is the line that moves EBITDA most, and inside gross profit is the line that keeps it from moving further.

    How a feasibility study turns the drivers into a forecast

    Everything above is input. What a lender receives is projected gallons, fuel margin, inside sales and EBITDA for each of the first five operating years, and the method that connects them is where most gas station studies fail the file. No federal agency, SBA procedure or Appraisal Institute text prescribes a per-capita benchmark, a capture rate or a long-term volume decay for fuel retail. That leaves the method to the analyst, and the analyst has to show it.

    Build gallons two ways and reconcile them. The demand-side method starts from the trade area: population, plus daytime workers and an adjustment for through traffic, times per-capita gallons from the state energy data, adjusted down for density, transit and remote-work share; the site’s share is then set by its fueling positions against every competitor, weighted for price position, access, brand and food program (10, 19). The supply-side method starts from the road: usable two-way traffic, times a capture rate that falls as traffic rises, times 8.4 gallons per transaction (12, 19). Both are then tested against the actual gallons of named comparable stations in the trade area. When the two methods disagree by more than 15 to 20 percent, the study explains the gap rather than averaging it away, and the capture rate it settles on must be consistent with the pass-by share in the traffic impact study.

    Count every competitor, including the ones not yet built. The competitive census lists each station by fueling positions, brand, price tier, food program and hours, including hypermarket and club fuel within about seven miles and every permitted or announced site. A planned competitor within a mile gets an explicit price haircut of about 2.5 to 2.7 cents a gallon, scaled up toward 5 cents for a hypermarket, plus a separately stated volume-diversion assumption (22). In a spacing or ban jurisdiction, the study says so and credits the subject with whatever protection the ordinance provides (35).

    Model the margin, not just the volume. Fuel margin gets its own base and downside, bracketed by the 2014 to 2019 range of roughly 22 to 24 cents and the 2024 to 2025 level of 39.7 cents, with card fees modeled per gallon rather than buried in a percentage (23, 11). The subject’s expected posted price against its three nearest competitors and the nearest club is stated, and the volume consequence of that position is shown as a derived sensitivity using a disclosed elasticity band scaled to local station density (21).

    Forecast the store as two businesses. Merchandise and foodservice are projected separately, each with its own margin. Inside sales are triangulated three ways: inside sales per gallon from format-matched comparables, sales per square foot from quartile-appropriate benchmarks with the unit confirmed, and a transaction check of traffic times conversion times basket, using regional baskets rather than the national figure (13, 15, 17). Foodservice is tested against kitchen size, staffing, drive-through stacking and a ramp of its own. The study underwrites to the operator’s model, not the NACS average, because same-store results across chains currently range from about +4 percent to −4 percent (16, 36).

    Show the ramp and the decay. Year one is not the stabilized year. Fuel ramps to roughly two-thirds to three-quarters of stabilized volume in year one for an independent, and merchandise more slowly, consistent with the operator evidence on new stores (15, 16). After stabilization, per-capita gallons erode about 1 percent a year in the base case and 2.5 percent in the downside, from a base year normalized for the 2026 price shock (2, 3, 4). Diesel is driven by freight and industrial activity, not population, and is projected separately.

    Tie the long-term view to value. In high-EV markets the study carries the state registration overlay through to the remaining-economic-life conclusion. In every market it reports the going-concern value allocation the lender will need and separates the real estate from the business. A study built this way produces a forecast a credit officer can test line by line. The next section shows what that test looks like.

    From forecast to DSCR

    Debt service coverage is where the drivers stop being an industry discussion and become a credit decision. The ratio is simple: cash flow available for debt service divided by required annual principal and interest. What goes into each side, and which floor applies, depends on the program.

    SBA sets a hard floor. Under the SOP 50 10 8 series, a standard 7(a) applicant’s debt service coverage must be at least 1.15 to 1 on a historical or projected basis, with operating cash flow defined as EBITDA and debt service as the required principal and interest on all business debt, including the new SBA loan (38). Since March 1, 2026, 7(a) Small loans of $350,000 or less must clear 1.10 to 1 (38). Where affiliates are involved, a global analysis is required. SOP 50 10 8.1 governs loans receiving an SBA loan number on or after October 1, 2026, and the current text should be checked for any change to these thresholds before a file is finalized (38). Most lenders set their own targets above the SBA floor, commonly 1.20x to 1.25x or higher for special-purpose collateral.

    Gas stations carry program-specific requirements beyond coverage. SBA treats gas stations, with or without a convenience store, as special-purpose property, which raises the 504 borrower contribution to 15 percent, or 20 percent if the business is also new (38). The maximum 7(a) term for real estate is 25 years. The going-concern appraisal must allocate value among land, improvements, equipment and intangibles. Every gas station loan requires at least a Phase I environmental site assessment and documentation of tank and equipment compliance, with a Phase II where the Phase I recommends one (38). SBA does not require a feasibility study on every loan; regulation lets it request one, and the 504 credit memo guidance lists market saturation by industry and location, a unique concept, a highly specialized property, a project disproportionate to the community it will serve, and rapid growth as triggers (38). A new fuel site on a corridor with three stations already in the mile touches the first trigger directly.

    USDA leaves the coverage minimum to the lender and makes the study mandatory more often. Under 7 CFR Part 5001, a Business and Industry guaranteed loan above $1 million to a new business requires a feasibility study by an independent qualified consultant, with its content set by Appendix A to Subpart D (39). The rule does not impose a blanket coverage minimum for new projects; the lender sets one and writes it into the loan agreement. Where refinancing is the majority purpose for an existing business, the borrower must show historical coverage of at least 1.1 times on the proposed debt service, and applications involving affiliates or guarantors require a global analysis (39). Going-concern value is deducted from reconciled market value before the collateral discount, so the business value a gas station appraisal reports does not count as collateral (39).

    A worked example shows how thin the margin for error is. Take the stabilized P&L from the profitability section: EBITDA of $405,600. Assume a $3.6 million project financed with an SBA 7(a) loan of $3.06 million, 85 percent of cost, amortized over 25 years at an illustrative 9.5 percent. Annual debt service is $320,821, and stabilized coverage is 1.26x, comfortably above the SBA floor and at the bottom of most lenders’ comfort range. The table below holds debt service constant and changes one driver at a time.

    CaseWhat changesEBITDADSCR
    Stabilized baseAs modeled$405,6001.26x
    One competitor within a mileStreet margin down 2.7¢ on every gallon$365,1001.14x
    Year 10, base demand decayGallons down 1% a year for ten years$368,9001.15x
    Margin normalizes partwayFuel gross margin at 30¢$345,6001.08x
    Price-position or club lossGallons down 10%, inside sales down 5%$339,5001.06x
    Year 10, downside demand decayGallons down 2.5% a year for ten years$319,7001.00x
    Combined moderate downside30¢ margin, gallons down 5%, inside down 3%$312,8000.97x
    Margin back to 2014–2019Fuel gross margin at 24¢$255,6000.80x
    Year one of operations75% of stabilized gallons, 65% of inside sales$105,8000.33x

    Three findings follow. First, the street margin is the swing variable. One competitor’s price response alone takes this site below the SBA floor, and a return to the pre-2020 margin takes it below breakeven, with gallons unchanged. Second, coverage at stabilization says little about year one or year ten. The year-one shortfall against debt service is about $215,000, which a funded reserve or the sponsor’s liquidity has to carry, and the year-ten base case sits exactly on the 1.15x floor with no further erosion allowed. Third, the store is the cushion. Every case above holds inside sales flat or lower; a site whose inside gross profit grows even 1 percent a year in real terms more than offsets the base-case gallon decay, and a site whose store is flat or shrinking does not.

    Breakeven gallons is the cleanest single test. Solving the P&L for the gallons that just cover a required coverage ratio gives:

    ```latex \text{Breakeven gallons} = \frac{\text{Operating expenses other than fuel card fees} + k \times \text{Debt service} - \text{Inside gross profit} - \text{Other income}}{\text{Fuel margin per gallon} - \text{Card fee per gallon}} ```

    where k is the required coverage. For the example site, breakeven is about 1.17 million gallons at 1.00x, 1.36 million at 1.15x and 1.48 million at 1.25x, against a projection of 1.5 million. The site can lose 22 percent of its projected gallons before it stops covering debt service, but only 10 percent before it fails the SBA test and about 1 percent before it fails a 1.25x lender covenant. That cushion, stated as a percentage of projected gallons at each coverage level the lender cares about, belongs on the first page of every gas station feasibility study.

    What this means for loans being underwritten now

    Six conclusions follow for any gas station feasibility study written in the fourth quarter of 2026.

    The fuel margin should be underwritten below the 2024 to 2025 level, with a downside at the pre-2020 range. A 40-cent gross margin has held for four years, and on the example site it is the difference between 1.26x and 0.80x. A study that capitalizes it as permanent has made the biggest assumption in the file without testing it.

    Gallons should erode after stabilization. Every official forecast case has gasoline lower in 2050 than in 2025, and fleet efficiency alone takes about 1 percent a year off per-capita demand. Flat gallons in years three through ten are not a base case; they are an upside case that has not been labeled.

    The store should be forecast as the business it now is. Inside sales and foodservice produce most of the industry’s gross profit, and they are the only lines that can offset gallon decay. That means separate merchandise and food projections, format-matched ratios, a capacity test on the kitchen and an operator-specific growth rate rather than the NACS average.

    Competition should be quantified, not described. The one causal, U.S. measurement available, a 2.7-cent price response within a mile of a new entrant and about 5 cents for a hypermarket, is enough on its own to move a thin file below the SBA floor. Every permitted or announced competitor belongs in the model with that haircut and a stated volume assumption.

    Traffic should screen, not forecast. Thresholds and capture rates are rules of thumb; the one published regression explains about half the variation in capture, and 8.4 gallons per transaction is the only anchored conversion. Comparable-station gallons should govern the volume conclusion, and the capture rate must agree with the traffic impact study.

    Coverage should be reported at three points, not one: year one, stabilization and year ten, under base and downside demand decay, with the breakeven-gallon cushion stated at every coverage level the lender uses. A study that shows only stabilized coverage has answered the easiest question in the file.

    The purpose of a gas station feasibility study is not to confirm the sponsor’s gallon projection. It is to tell the lender which part of the cash flow is bought with a commodity margin set by the street, which part is earned inside the store, how much of each the trade area can deliver, and how far either can fall before the loan stops covering itself.

    Questions lenders and borrowers actually ask

    What drives gas station profitability more, gallons or the store?

    At the industry level, the store. In 2025 fuel was 65 percent of the average convenience store’s sales but 38.8 percent of its gross profit, and merchandise and foodservice produced 57.4 percent (11). At a given site the answer depends on format: a fuel-led, small-box site may still earn close to half its gross profit at the pump, which makes it more exposed to the street margin.

    What fuel margin should a feasibility study use?

    Not the 2024 to 2025 level as a permanent base. OPIS’s U.S. gasoline gross margin was 39.7 cents a gallon in both years, against roughly 22 to 24 cents between 2014 and 2019 (23). A defensible study runs a base case between the two, a downside at the pre-2020 range, and models card fees of about 8 cents a gallon separately (11).

    How many gallons does a typical gas station sell?

    There is no typical figure. NACS reports 322 fueling transactions a day at 8.4 gallons each, about 2,700 gallons a day, while other NACS publications cite 2,500 to 4,000 gallons a day on different bases (12). Murphy USA averaged about 236,000 gallons per store per month in 2025 and Casey’s about 101,000 per mature store per month in fiscal 2026 (15, 16). Independents should be benchmarked against comparable local stations and the lower industry quartiles, not the national mean.

    What capture rate should be applied to traffic counts?

    No published standard exists. The one published regression shows capture falling as traffic rises, from about 3.1 percent of passing vehicles at 10,000 a day to under 1 percent at 50,000 (19). Restate any capture assumption as daily fuel transactions over two-way adjacent traffic, check it against the traffic impact study’s pass-by share, and let comparable-station gallons govern the conclusion.

    What DSCR does SBA require for a gas station loan?

    A standard 7(a) loan must show at least 1.15x on a historical or projected basis, with operating cash flow defined as EBITDA; 7(a) Small loans of $350,000 or less must show at least 1.10x since March 1, 2026 (38). Lenders usually set higher internal targets for special-purpose collateral. A 504 loan on a gas station requires a 15 percent borrower contribution, or 20 percent for a new business (38).

    Does a gas station loan need a feasibility study?

    For SBA, not automatically; SBA may request one, and market saturation and a project disproportionate to its community are among the listed reasons (38). Many lenders require one for new construction under their own credit policy. For a USDA Business and Industry guaranteed loan above $1 million to a new business, the study is mandatory and must be prepared by an independent qualified consultant (39).

    How should EV risk be reflected in a gas station loan?

    Through the long-term volume assumption and the remaining economic life, not a blanket haircut. EVs were about 6.4 percent of new sales in the third quarter of 2026 and about 2 percent of vehicles on the road, and efficiency, not EVs, is driving most of today’s gallon decline (26, 27, 2). In California, Washington, Colorado, Oregon and the New Jersey and D.C. corridor, a 25-year amortization should be justified against local registration trends.

    What happens to a station when a Buc-ee’s, Wawa or warehouse club opens nearby?

    The measured effect is on price: incumbents within a mile cut about 2.7 cents a gallon after a typical entry and about 5 cents after a hypermarket opens (22). No public study measures the volume lost. A study should apply the price effect directly and state its volume-diversion assumption, because on a thin file the price effect alone can move coverage below the SBA floor.

    Related research and case studies

    The drivers above are applied, with the fuel margin normalised and the store and fuel lines underwritten separately, in our gas station acquisition case study in Gettysburg, Ohio. The same coverage discipline appears across our other case studies, including the express car wash in North Port, Florida, the collision repair acquisition in Bensalem, Pennsylvania, the hotel acquisition in Alice, Texas, the new hotel development in Gaffney, South Carolina, the RV park acquisition in Wolfeboro, New Hampshire and the marina acquisition in Fort Myers Beach, Florida.

    For more on the asset class, read our guides to gas station site selection, traffic counts and road speed, gas station operating expenses and the pro forma, the convenience store industry in 2026 and the future of foodservice, EV exposure in gas station feasibility, gas station zoning, regulatory requirements for gas station development, finding a fuel supplier, appraising gas stations and truck stop operating expenses. For the methods, see the 2026 Feasibility Methodology Atlas, DSCR in feasibility studies, sensitivity analysis, operating expense ratios and cap rates by asset class, going-concern allocation and how USDA treats going-concern value for hotels, gas stations and car washes. On the regulatory side, see Three Reports, One Credit File, the 7 CFR Part 5001 Handbook and appraisal requirements by loan program.

    If you need these drivers tested against a specific site, UnionMetric prepares lender-ready gas station feasibility studies for SBA and USDA financing.

    Sources

    1. U.S. Energy Information Administration, FAQ “How much gasoline does the United States consume?” (2018 record) and Gasoline Explained: Use of Gasoline (2025 volume); Petroleum Data, status of the suspended Adjusted Sales of Fuel Oil and Kerosene series
    2. U.S. Energy Information Administration, Today in Energy, April 10, 2026 (2025 gasoline consumption, VMT and implied fleet fuel economy; fuel economy assumption)
    3. U.S. Energy Information Administration, Short-Term Energy Outlook, July and October 2026, including Table 4a (consumption forecasts and retail prices)
    4. U.S. Energy Information Administration, Annual Energy Outlook 2026, released April 8, 2026, Narrative and case descriptions; 2050 motor gasoline values by case as computed from the AEO2026 data file by a secondary source
    5. Federal Highway Administration, Traffic Volume Trends, December 2025, and Highway Statistics 2024, Table VM-1 (updated February 2026)
    6. U.S. Environmental Protection Agency, The 2025 EPA Automotive Trends Report, EPA-420-S-26-001, February 2026
    7. National Highway Traffic Safety Administration, The SAFE Vehicles Rule III for Model Years 2022 to 2031, Federal Register, September 30, 2026; Congressional Research Service, R49361
    8. U.S. Environmental Protection Agency, Rescission of the Greenhouse Gas Endangerment Finding and Motor Vehicle Greenhouse Gas Emission Standards, 91 Fed. Reg. 7686, February 18, 2026; Congressional Research Service, R49024
    9. U.S. Census Bureau, American Community Survey 2024 1-year estimates and 2019–2023 releases, work-from-home and commuting mode shares, as compiled by secondary sources
    10. U.S. Energy Information Administration, Today in Energy, August 31, 2020 (per-capita motor gasoline by state, 2018) and Today in Energy on state gasoline expenditures; State Energy Data System 2024, Table F15
    11. NACS, State of the Industry data for 2025, press release of April 15, 2026, and NACS Magazine and NACS Daily coverage of the 2024 and 2025 State of the Industry reports, including foodservice data
    12. NACS, Key Facts About Fueling (updated June 29, 2026) and Consumer Fuels Survey results reported there
    13. NACS Magazine, regional reports from the 2026 State of the Industry Summit, June 2026, and prior regional reports, June 2024 and June 2025
    14. NACS and TEI, Consumer Fuels Survey 2025 (fielded May 19–21, 2025) and prior NACS consumer fuels surveys
    15. Murphy USA Inc., Form 10-K for 2025, fourth-quarter 2025 results and prepared remarks, and investor presentations filed on Form 8-K (2020, March 2022 and March 2026)
    16. Casey’s General Stores, Inc., Form 10-K for fiscal 2026 and fourth-quarter fiscal 2026 results release; earnings call commentary and investor presentations, fiscal 2025–2026
    17. Yesway, Inc., Form S-1/A registration statement, 2026
    18. Wawa, Sheetz and Refuel Market, published real estate site selection criteria (current web pages)
    19. Spack Consulting, as published in CStore Decisions, “Selecting an Optimal Location for Your C-Store,” September 15, 2017; Robert E. Bainbridge, “Site Essentials of Convenience Stores and Retail Fuel Properties,” The Appraisal Journal, Winter 2012; Bainbridge, Convenience Stores and Retail Fuel Properties: Essential Appraisal Issues, 2nd ed., Appraisal Institute, 2012
    20. Institute of Transportation Engineers, Trip Generation, pass-by rates for convenience stores with fuel, as compiled by Wisconsin DOT; North Carolina DOT, “Traffic Operational Impacts of Contemporary Multi-Pump Island Fueling Centers,” 2007; Texas Transportation Institute, Report 3904-5, 2000
    21. Barron, Umbeck and Waddell, “Consumer and competitor reactions: Evidence from a field experiment,” International Journal of Industrial Organization 26, 2008; Wu, “Search with learning in the retail gasoline market,” RAND Journal of Economics, 2024
    22. Taylor and Muehlegger, “The Effects of Competition in the Retail Gasoline Industry,” NBER Working Paper 33569 (March 2025, revised April 2026), Federal Reserve Bank of Dallas Working Paper 2509, and Journal of the Association of Environmental and Resource Economists 13(5); Davis, McRae and Seira, Journal of Industrial Economics, 2025 (Mexico); retail entry study, Journal of the European Economic Association (Germany)
    23. OPIS, MarketSharePro and MarginPro data as published by OPIS and NACS, 2018–2025 (visit shares, branded and unbranded efficiency, retail gasoline margins, store expense trend); CNN Business, November 20, 2021, quoting OPIS on Costco volumes
    24. Costco Wholesale Corporation, Form 10-K for fiscal 2025; BJ’s Wholesale Club Holdings, second-quarter fiscal 2026 earnings call
    25. Verifone, fuel loyalty transaction data release, July 2, 2026, and summer 2026 data; Cumberland Farms, Form F-1 registration statement, 2026
    26. Cox Automotive and Kelley Blue Book, EV Sales Reports for the second and third quarters of 2026 and Q3 2026 Sales Forecast Update
    27. Experian, Automotive Market Trends and Automotive Consumer Trends, second quarter 2026
    28. BloombergNEF, Electric Vehicle Outlook 2026, June 16, 2026; International Energy Agency, Global EV Outlook 2026
    29. Public Law 119-21 and IRS guidance on the termination of the 30D, 25E, 45W and 30C credits; Public Law 119-16 and related litigation, including the September 2, 2026 preliminary injunction (D.D.C.); Congressional Research Service, R48996 (NEVI); National Conference of State Legislatures, EV registration fees; U.S. DOE Alternative Fuels Data Center, state EV registrations, 2025; Alliance for Automotive Innovation, Get Connected, Q3 2025
    30. Paren, U.S. DC fast-charging reports, second and third quarters of 2026
    31. Matthews, Convenience Store Market Report, November 2025; The Boulder Group, Net Lease Market Report, third quarter 2026
    32. Morningstar DBRS, 2026 North American Fuel Retail and Convenience Store Outlook
    33. NACS and NIQ TDLinx, 2026 Convenience Industry Store Count (stores open December 31, 2025) and prior annual counts
    34. Transaction disclosures: Federal Trade Commission consent order on 7-Eleven and Speedway, 2021; BP release of February 16, 2023 on TravelCenters of America; Casey’s Form 8-K of July 26, 2024 on Fikes/CEFCO; Alimentation Couche-Tard releases on GetGo, 2025, and letter of July 16, 2025 to the Seven & i board; Sunoco LP Form 8-K of May 5, 2025 on Parkland
    35. Municipal records and local reporting: Petaluma, Santa Rosa, Windsor and Sonoma County, California ordinances, 2021–2023; Kansas City, Missouri Ordinance 240157, 2024; Rome, Georgia gas station regulations, January 2026; Kokomo, Indiana spacing ordinance; Denver Bills 241866 and 241867 (secondary report); Chicago Ordinance O2025-0018183 (pending)
    36. ARKO Corp., fourth-quarter and full-year 2025 results
    37. U.S. Environmental Protection Agency, 40 CFR Part 280 as revised July 15, 2015 (80 Fed. Reg. 41566), and UST Performance Measures, September 2025 and March 2026; California State Water Resources Control Board, single-walled UST closure program under SB 445
    38. U.S. Small Business Administration, SOP 50 10 8 (effective June 1, 2025) and SOP 50 10 8.1 (loans numbered on or after October 1, 2026); Procedural Notice 5000-846607 (2023) and the 2026 procedural notice revising 7(a) Small loan underwriting effective March 1, 2026; 13 CFR 120.160(b) and 120.910; SOP appendix requirements for gas station loans, as quoted in secondary sources
    39. 7 CFR Part 5001 (OneRD Guaranteed Loan regulations), including the refinancing, credit evaluation and feasibility study provisions and Appendix A to Subpart D; USDA Rural Development, Business and Industry lender application checklists
    40. SBA 7(a) and 504 loan-level FOIA data as tabulated by third parties, including the PeerSense 2026 SBA Lending Report and published cohort studies of fiscal 2010–2016 vintages; secondary
    41. C-Store Dive, report on NACS CSX Convenience Benchmarking data presented at the NACS Show, October 7, 2026
    42. Retail Petroleum Consultants, presentation to the Society of Chief Appraisers, 2015 (going-concern valuation conventions)

    Related feasibility study engagements

    RM

    Robert Mitchell, MAI, SRA, AI-GRS

    Certified commercial real estate appraiser with extensive experience in industry guides property valuations across the United States.

    View team credentials →
    Get Started Today

    Need to Know if Your Project Works?

    Tell us about your project and we'll send a clear proposal: scope, timeline, and fee.