read white paper


July 2023

EVs and biofuels reshape US oil demand outlook

US gasoline consumption peaked two decades ago and has stayed broadly unchanged since then (whilst GDP has grown from $10trn to $23trn!). More notably, we now believe we are on the cusp of an accelerating decline. The gasoline demand decline will more than offset growth from jet fuel and trucks, hence driving overall US oil consumption lower.

In this quarter’s Special Section, we take a closer look at the impact that fuel efficiency regulations, Electric Vehicles (EVs) and biofuels are having on US oil product demand, and therefore by consequence, oil demand. We also examine why previous oil demand growth trajectories (particularly those tied to GDP growth) will no longer hold as consumption pattern changes take hold. We conclude on the implications for the US refining sector, as well as broader portfolio implications.

Why is this relevant now? Most recently, on 12th April, 2023, the US Environmental Protection Agency (EPA) announced new, more ambitious proposed standards to further reduce harmful air pollutant emissions from light-duty and medium-duty vehicles starting with model year 2027. The proposal builds upon EPA’s existing standards for passenger cars and light trucks for model years 2023 through 2026. The proposed standards, which would phase in over model years 2027 through 2032, are significantly stricter than those preceding them.

Whilst the focus of our work here is assessing implications for gasoline and diesel markets in the US, we believe this research could offer a window into how quickly demand trends can change more globally. A subject for a future letter.

US gasoline demand

The US is electrifying passenger cars at increasing pace, with a further catalyst offered via the recently passed Inflation Reduction Act (IRA), we think this is significant for gasoline demand.

A structural decline in US gasoline consumption will reverberate globally. This is because Americans drive a lot. Americans drive on average 2-3x more per capita than most other OECD countries. As a result, the US consumes more gasoline than any other country, accounting for 34% of world’s gasoline consumption and 19% of global oil demand. But gasoline consumption in the US peaked in around 2006-2007, at around 9.4 million barrels per day (mbd) when miles travelled were 5% lower than today. Why is that?

US Gasoline Demand has been impacted by several factors:

-       efficiency measures (cars and light trucks getting increasingly more fuel efficient).

-       structural changes (work from home trends).

-       increasing ethanol blending. 

Indeed, we estimate the above-mentioned factors have impacted demand for gasoline by around 0.5 mbd, split predominantly across efficiency measures and structural changes. By 2030, we calculate we could lose another 1.4 mbd of gasoline demand predominantly through efficiency measures. 

Motor gasoline (incl. ethanol) has been broadly flat for over 20 years

motor gas

Source: US Department of Transportation, Department of Energy, Clean Energy Transition analysis

In our analysis we focus on two of the three factors: namely motor vehicle efficiency and ethanol blending: both broadly similar in demand impairment thus far, but the former will make more of an impact going forward.

Vehicle efficiencies show their impact

If we were to extrapolate miles driven trends, gasoline consumption “should” currently stand closer to 9.1 mbd (from 8.6 mbd currently), therefore, trends in fuel efficiencies have caused around 11.5% of gasoline demand from transportation to “disappear”. We can see this in the miles driven per unit of gasoline chart below. On average, the US driver is getting 18% further per unit of petroleum gasoline in 2023 than in 2000, broadly corroborating the “missing demand” picture described above. We explore this impact further in the Special Section, as the impact of the recent EPA proposal could significantly accelerate the efficiency statistics presented here.

18% improvement in miles driven per unit of petroleum gasoline since turn of decade


Source: US Department of Transportation, Department of Energy, Clean Energy Transition analysis

EPA forcing automakers to electrify

While the fuel efficiency of cars increased tremendously during the 1970s as a response to surging fuel costs, this development stopped and partly reversed between 1985 and 2005. While there was underlying improvement in engine technology, low oil prices and limited regulation meant that automakers and consumers gradually shifted to bigger cars with large gasoline engines, paying little attention to the impact of fuel consumption.

Fuel consumption of new vehicles


Source: EPA, Clean Energy Transition analysis

This changed in 2007 when the EPA started to phase in more stringent fuel efficiency regulations for the auto industry. The development was however somewhat slowed down during the Trump administration, but again resumed during the Biden administration. From model year 2022 to 2032 the EPA targets are becoming increasingly aggressive, requiring radical improvement in average fuel consumption for new vehicles. From 2022 to 2027 US automakers need to cut emissions by 32% for new car sales. The US EPA rules are very stringent, and avoiding compliance is not a real option, as any breaches could have devastating financial consequences for the automakers.

Figure 8 shows the average real-world emissions of new cars in the US (light blue) and the average real-world emissions of the fleet (dark blue). The data from 2023 to 2032 is based on the average automaker achieving the bare minimum of the legal requirement, hence represents a reasonably conservative scenario. Given that American passenger vehicles consume gasoline, CO2 emissions can be translated directly to gasoline fuel consumption.

Average US car emissions


Source: EPA, Clean Energy Transition analysis

While automakers are continuing to improve the energy efficiency of their cars, the only real option to hit the new stringent rules is to increasingly shift their car portfolio to battery electric. To hit the 2032 targets 60% of new car sales in the US need to be EVs. While this might sound like a high number, it is consistent with automakers own production plans. The EV shift will of course also be helped by the IRA subsidies. Given the above is a legal minimum, there is some modest upside risk to EV penetration.

Assuming vehicle miles travelled grow in line with the trend we saw pre COVID, the net result is that US gasoline demand is already in structural decline. In our view, energy efficiency and electrification are resulting in gasoline demand currently declining by 1.5-2% per year. This trend will accelerate in the second half of the 2020s towards 4-5% per year as the EPA emission regulations tighten. We forecast that by 2030 US gasoline demand will be 25% lower than 2019, and 17% lower than 2023. Over the next 7 years demand for gasoline will shrink by 1.4 mbd.

US gasoline demand projected to fall significantly into 2030s (ex-ethanol)


Source: EIA, Clean Energy Transition analysis

Ethanol content in gasoline impacting petroleum demand

One seldom discussed impact on petroleum demand is the impact of bio and renewable fuels. When it comes to gasoline, over the last 2 decades there has been a significant step up in blending of ethanol into finished motor gasoline, and this has stabilised close to 10% of all gasoline fuel consumed in America.

The EPA has approved use of E15 (15% Ethanol) in all vehicles made in 2001 and beyond, and according to the US Department of Agriculture, approximately 93% of motor vehicles registered on the road in the US can use E15. Some states (e.g. Minnesota) have seen Ethanol comprising over 12.5% of total gasoline consumed. The oil industry has frequently flagged the “blend wall” arguing that ethanol's percentage of fuel consumed should be capped at 10% (E10), so as to not harm the fuel systems and engines of cars not designed to run such a high proportion of ethanol. For the purposes of our estimates, we assume ethanol blend stays at 10%; any upside to that blend % would offer headwinds to petroleum gasoline demand.

% of ethanol blended into finished gasoline has grown significantly since start of the Millenium

% of ethanol blended into finished gasoline has grown significantly since start of the Millenium

Source: US Department of Transportation, Department of Energy, Clean Energy Transition analysis

In summary, our research leads us to conclude that:

-       Americans are driving more on less gasoline. 

-       A confluence of three factors likely contributed to the decline: vehicle fuel efficiencies, miles driven trends e.g., WFH (potentially somewhat structural) and ethanol blending.

-       Stringent EPA targets forces automakers to electrify, resulting in close to 50% of all new vehicles being electric by 2030.

-       Under such a scenario, US gasoline consumption will likely slump by about 17% between 2023 and 2030 to 7.1 mbd, for a total decline of 1.4 mbd over the period, despite Americans driving 8.4% more miles between now and 2030.

-       Given gasoline’s 44% share in total US oil demand, US oil consumption is poised to eventually contract between now and 2030, but by 1.3 mbd with gasoline losses being the main driver; small declines in diesel offset by jet fuel and petrochemicals growth.

-       A structural decline in US gasoline consumption will reverberate globally given that the US consumes more gasoline than any other country, accounting for 34% of world’s gasoline consumption.