An Analysis of The Valero Benicia Refinery Closure on Gasoline Prices in California

In April, Valero announced its Benicia, California refinery will cease operations in 2026. This surprise announcement has generated concern among policymakers that gasoline prices in the state may permanently shift higher. In an analysis with my SIEPR colleague (and PhD advisee) Ryan Cummings, we find that the price effects at the pump from the closure are likely between negligible—because the marginal barrel is already imported—and 15 cents/gallon as a conservative upper bound. Importantly, this is conditional on streamlined permitting which allows the refinery to convert to a product terminal.

Summary

This analysis examines the feasibility of replacing lost gasoline production capacity from the closure of Valero’s Benicia refinery with marine imports from the global market.

1. An analysis of potential bottlenecks shows there is sufficient global refining capacity and sufficient in-state port, storage, and distribution capacity to replace lost in-state production with marine imports, conditional on updating port infrastructure to facilitate gasoline imports. In order to make these updates in a timely manner, streamlined permitting may be necessary.

2. A preliminary analysis indicates that the cost impacts of switching to marine imports could be negligible and would be no more than 15 cents/gallon as a conservative upper bound. To prevent further consolidation in the market, Benicia’s storage and distribution infrastructure should not be acquired by an incumbent with substantial in-state market share.

   1. Potential, future refinery closures do not change the underlying economics on how the price of wholesale gasoline is being set in California. Because of this, additional refinery closures–conditional on logistical bottlenecks being cleared–would not increase or change these price impact estimates.

3. An alternative approach of subsidizing Valero to keep Benicia open could result in significant overpayments and generate a vicious cycle where additional financially viable refineries threaten shutdowns to extract their own subsidies. Labor and local economic concerns are better addressed through targeted transfers to affected workers and the local community.

Background on Supply-Demand Imbalance

California’s daily gasoline demand averages 887,000 barrels/day, roughly 120,000 barrels/day in excess of the state's 760,000 barrels/day in production.^[1] The gap is currently filled by marine imports of gasoline (and blending components) from refineries largely located in Asia and the Middle East.^[2] The recently announced closure of Valero’s Benicia refinery, which provides CA with roughly nine percent of its refining capacity, would increase the gap between in-state production and consumption by roughly 75,000 barrels/day to a total of 195,000 barrels/day. While California’s gasoline demand is expected to decline by as much as 40% over the next 25 years, refinery closures that outpace demand reductions may expose California to ongoing and significant gaps along this transition path.

Potential Bottlenecks to Increased Marine Imports

Replacing in-state production with marine imports presents three potential bottlenecks:

● Global CARBOB refining capacity: Global CARBOB gasoline and blending component capacity needs to be sufficient to replace the lost in-state production.

● Port throughput: California’s port system — including product terminals and ports located within refineries – needs to have sufficient throughput to handle the increased import volume.

● Storage and distribution: California’s storage and distribution infrastructure needs to have sufficient capacity to accommodate lumpy imports and distribute gasoline to consumers.

Based on analysis and conversations with other experts, we conclude that these potential bottlenecks will not prevent marine imports from offsetting lost in-state production, conditional on updating port infrastructure to facilitate gasoline imports. We discuss each below.

Global CARBOB capacity. Gasoline sold in California must meet the specifications laid out by the California Air Resources Board (CARB). CARB-compliant gasoline, without ethanol, is referred to as “California Blendstock of Oxygenated Blending,” or CARBOB. While some foreign refineries produce finished CARBOB, California mainly imports “blendstocks,” which are components of gasoline that are mixed together to make gasoline that meets CARBOB specifications. This mixing, called “mechanical blending,” is then done at an in-state blending plant, typically located at a fuel terminal or in a refinery.^[3],[4] The blendstocks that are used to meet CARB specifications are produced by many facilities worldwide, and global refining capacity is expected to increase by roughly 2.5 mb/d by 2030.^[5] For most blendstocks, the incremental California demand is small relative to the size of the global market. Given enough time for global facilities to adjust their production, there is unlikely to be bottlenecks for global CARBOB and CARBOB blending component production.^[6]

Port throughput. CARBOB and blending components can be offloaded from marine vessels at product terminals, such as the Chevron Richmond refinery or at the San Pedro Bay berths in Los Angeles. In addition, gasoline products can be sent directly from marine vessels to within-refinery storage tanks, from which they can then be sent into the regular distribution network.

The main constraint faced at ports is the time to offload the product. Because product pipelines tend to be smaller than crude pipelines, less product can be piped into storage per unit time. Conversations with distribution and logistics experts indicate that it currently takes 4-5 days to offload a 300,000 barrel MR tanker at the Benicia refinery. This means that roughly 60,000-75,000 barrels/day of gasoline could be piped in on a daily basis, which is at the low end of the lost Benicia production.

There are several options to mitigate this potential bottleneck. Benicia could be transformed into a product terminal, converting its more capacious crude pipelines and storage to be used for gasoline. This would allow upwards of 125,000 barrels/day to be offloaded into storage, more than enough to fill the gap created by Benicia’s closure. The potential bottleneck could alternatively be addressed by infrastructure updates to the Richmond Kinder Morgan product terminal or by using the dock at the PBF Martinez refinery. Converting Benicia to a product terminal or updating the infrastructure at the Richmond Kinder Morgan product terminal likely requires separate permitting processes with the city, county, State Lands Commission, Air Quality Management District, among other entities. Provided the permitting could be completed in a timely manner, marine import capacity will be sufficient to cover the increase in imports.^[7]

Storage and distribution. Because the additional marine imports would arrive on large tankers on a periodic basis, reliance on imports may require additional use of in-state storage. However, California has more than enough storage capacity to handle these lumpy deliveries. The state currently possesses roughly 24 million barrels of storage for refined products, which includes storage for other products such as diesel, jet fuel, and ethanol.^[8] Of that, roughly three-quarters is dedicated to gasoline, of which two-thirds or 12 million barrels of capacity within the refinery gate.^[9],[10] Additional gasoline storage capacity could be obtained by converting the crude storage tanks at Benicia to product tanks, as was recently done at the Rodeo and Martinez refineries.

The distribution network from storage facilities to the end customer will be unaffected by the Benicia refinery closure and therefore does not pose any bottleneck risk.

Cost of imports. The incremental cost of switching to marine imports from in-state production to imports depends on the cost differential for the replacement refining capacity and the incremental transport costs of marine imports. Using data from publicly posted tanker freight rates, we estimate costs of roughly 13-15 cents/gallon to transport product to California from Asia.^[11] However, since the marginal barrel of gasoline in California is already imported, these transport costs may already be reflected in baseline prices. In addition, newer, foreign refineries that produce CARBOB blending components, such as those in India, have lower production costs than the Valero Benicia facility, which was built in 1968.^[12] As a result, we believe that the retail price effects experienced by consumers of replacing Benicia with maritime imports could be negligible and are no more than 15 cents/gallon as a conservative upper bound.

Competitive dynamics. Imports have historically been brought into California by the largest refiners in the state, such as Chevron. If the largest refiners are able to control incremental marine imports, they may be able to increase markups in the downstream spot and rack markets. To prevent further consolidation, Benicia’s storage and distribution infrastructure should not be acquired by an incumbent with substantial in-state market share. More generally, the state could seek to structure imports in a way that enhances market competition.

Economic security concerns. Economic security concerns of additional marine imports are limited. First, California refineries already import nearly two-thirds of the crude oil used to make gasoline from foreign countries, generating unavailable baseline risk exposure. Second, much of the refined product that is imported into California comes from U.S. allies such as Korea, India, and Singapore (which acts as a storage hub).

Generalizing the analysis. Because California will most likely continue to be a net importer, the economics described above will not change; the marginal, imported barrel will continue to set the price in California as it does today. As a result, additional refinery closures–conditional on logistical bottlenecks being cleared–should have limited effects on prices.

Logistical bottlenecks will, of course, need to be examined on a case-by-case basis. But it stands to reason that converting future shuddered refineries to product import terminals will also be a workable remedy to potential bottlenecks. For example, with the exception of a single refinery that produces a small amount of gasoline, every refinery in the state that produces gasoline also possesses their own marine-import terminals along with ample storage.^[13]

Policy Responses

An alternative approach is to subsidize Valero to keep the Benicia refinery open. This approach raises two concerns. First, asymmetric information between Valero and the state creates significant overpayment risk. Valero knows more about their operating costs than the state. If the state opts to subsidize, Velero will push for the largest possible subsidy, and the state may not have the information or risk appetite to push back. Second, if the state subsidizes Benicia, it may trigger a vicious cycle in which other financially stable refineries threaten to shut down in order to receive their own subsidies.

Labor and local economic concerns. Refineries in California currently employ 6,000-7,000 people directly, with roughly 400 workers at the Valero Benicia refinery.^[14] Income losses from well-paying, unionized refinery jobs and the spillover effects on the local economy are best addressed through targeted payments. For example, workers could be offered buyout contracts, and the local economy could be supported through grants. In addition, the conversion of Benicia to a product terminal would provide employment opportunities for refinery workers, as would regular operations of the product terminal.

---

^[1] The state also produces roughly 85,000 barrels/day of non-California gasoline, which is primarily exported via pipeline to Arizona and Nevada.

^[2] Due to Jones Act restrictions, there are only 55 product tankers in the U.S. capable of carrying refined product from one U.S. port to another – and these product tankers are often the highest-cost option. As a result, there are limited product transfers between California and the rest of the U.S., with the exception of Washington State. Washington State provides California with up to 45% of its imported supply when refineries are operating without disruptions, but this share often shrinks to less than 20% during refinery outages, in which case foreign supply fills the gap. See, e.g., https://efiling.energy.ca.gov/GetDocument.aspx?DocumentContentId=95154&tn=259097

^[3] See, e.g., https://krohne.com/en/industries/oil-gas-industry/refining-oil-gas-industry/gasoline-blending-in-the-oil-and-gas-industry

^[4] Ethanol is typically blended closer at a fuel terminal called “a rack,” where gasoline is put onto trucks before being delivered to retail stations.

^[5] See, e.g., https://www.iea.org/reports/oil-2025/executive-summary

^[6] For example, global alkylate capacity is roughly 2.3 mbpd, Alkylate is around 15% of CARBOB gasoline, thus California’s consumption of alkylate a little over three percent of global consumption (assuming the market is balanced).

^[7] Here, we do not discuss the impact of CARB at-berth regulations on marine imports, as our understanding is the program was extended earlier this year and the impact of the regulations is still uncertain.

^[8] See Table 3 here. The State’s A08 survey covers “tanks located at refinery and refiner-owned offsite storage facilities,” and amounts to roughly 20 million barrels. This does not include marine storage tanks, though. When including these, we include the Selby, CA NuStar tank (2.67M barrels, see p.11 here under “storage segment”) and the Zenith Alamitos terminal (2.61M barrels, see here) In our calculation of total storage, we omit pipeline breakout tanks, such as those owned by Kinder Morgan (see here for a list of pipeline breakout tanks), as those are for temporary storage that often sits less than 24 hours. Including breakout tanks would increase statewide storage capacity by roughly 2.5M barrels.

^[9] EIA data shows that in 2024, storage of gasoline and blendstocks in California averaged roughly 18M barrels, or 23 days of gasoline demand.

^[10] This is calculated by summing reported product storage capacity for each refinery, and assuming 35% is devoted to product storage, and further assuming that gasoline is roughly 65% of all products.

^[11] See, e.g., https://www.argusmedia.com/-/media/project/argusmedia/mainsite/english/images/downloadable-content/20240904freighttanker_sample.pdf. Argus does not regularly post U.S. to Asia freight prices for clean product tankers, so as a result, we look at routes with similar distances and calculate the cost

^[12] For example, the Benicia refinery was initially created to process Alaska North Slope (ANS) crude, and therefore has a limited degree of flexibility in terms of which inputs it is optimized for. Newer refineries in Asia and the Middle East, by contrast, are designed to process a wider variety of crudes. Furthermore, these refineries have substantially better temperature and pressure control, which generates much higher product yield. Experts we have spoken with said these features together lead to foreign refineries supplying CARBOB blending components of operating costs that are 10-15% lower than Benicia’s.

^[13] In particular, the Kern refinery is the only land-locked refinery that produces gasoline. This refinery, when operating, produces between 8,000-10,000 barrels/day of gasoline.

^[14] See https://www.valero.com/about/locations/benicia-refinery