The global market for liquid fuels (oil, biofuels and other liquids) transitions as oil demand peaks and supplies shift.
The demand for liquid fuels in Rapid and Net Zero never fully recovers from the fall caused by Covid-19, implying that oil demand peaked in 2019 in both scenarios.
The consumption of liquid fuels falls significantly over the Outlook in both scenarios, declining to less than 55 Mb/d and around 30 Mb/d in Rapid and Net Zero respectively by 2050. The falling demand is concentrated in the developed world and China, with consumption in India, Other Asia and Africa broadly flat over the Outlook as a whole in Rapid, but falling below 2018 levels from the mid-2030s onwards in Net Zero.
In contrast, after recovering from the impact of Covid-19, the consumption of liquid fuels in BAU is broadly flat at around 100 Mb/d for the next 20 years, before edging lower to around 95 Mb/d by 2050. Demand for liquid fuels continues to grow in India, Other Asia and Africa, offset by the trend decline in consumption in developed economies.
Despite the weakness in oil demand, US tight oil* in Rapid recovers from the impact of Covid-19 and expands until the early 2030s, with this increase in output more than offset by falls in OPEC production. Thereafter, OPEC production broadly stabilizes as declines in global demand are broadly matched by falls in US tight oil and other non-OPEC supplies. By 2050, non-OPEC supplies account for around two-thirds of the total decline in liquids supply in Rapid.
US tight oil also grows over the next 10 years or so in BAU offset by declining OPEC production. Declines in US tight oil and other non-OPEC output from the mid-2030s onwards provides scope for OPEC to increase its production despite the backdrop of gradually declining demand. By 2050, the level of OPEC production in BAU is broadly unchanged from its level in 2018.
The outlook for liquid fuels demand is dominated by the impact of Covid-19 in the near-term and by their declining use in the transport sector further out.
Global liquids demand in all three scenarios is significantly affected by the impact of Covid-19, which disproportionately impacts economic activity and prosperity in emerging economies which are the main growth markets for liquid fuels. The experience of coronavirus also triggers some lasting changes in behaviour, especially increased working from home (see Global backdrop).
Liquids demand in Rapid and Net Zero does not fully rebound from this near-term hit to demand and subsequently falls substantially over the Outlook to around 55 Mb/d and 30 Mb/d by 2050 respectively.
The scale and pace of these falls are driven primarily by the changing use of liquid fuels in the transport sector, which declines sharply in the second half of the Outlook in both Rapid and Net Zero, driven by the increasing efficiency and electrification of road transportation (see Transport). The transport sector accounts for around two-thirds of the decline in the use of liquid fuels by 2050 in Rapid and almost 60% in Net Zero.
The pace of decline in liquids demand in the second half of the Outlook – during which it falls by an average of over 2 Mb/d per annum in Rapid and 3 Mb/d in Net Zero – is unprecedented and has significant implications for other parts of the oil industry, including refining (see Oil).
Liquids fuel demand in BAU is more resilient, with broadly stable use in transport helping to support global demand at around 100 Mb/d for much of the next 20 years, before it edges lower in the final 10 years of the Outlook as the use of liquid fuels within transport begins to decline.
The non-combusted use of liquid fuels, largely as a feedstock in the petrochemicals sector, provides some degree of support to overall liquids demand, increasing in both Rapid and BAU, and declining below 2018 levels only in the final 10 years of the Outlook in Net Zero.
The composition of global liquids supply is initially dominated by a rebound in US tight , with OPEC’s share of production recovering in the second half of the Outlook.
In Rapid, US tight oil recovers from the falls caused by the impact of Covid-19, increasing to close to 15 Mb/d in the early 2030s. Brazilian production also grows over the same period. But as US tight formations mature and OPEC adopts a more competitive strategy against a backdrop of accelerating declines in demand, US production and non-OPEC output more generally fall from the early 2030s onwards.
OPEC production declines over the next 10 years or so before broadly stabilizing thereafter, with its share of total liquids production recovering from a low of close to 25% in the early 2030s to around 45% by 2050. The higher cost structure of non-OPEC production means around two-thirds of the total fall in liquids production in Rapid by 2050 is borne by non-OPEC supplies.
Non-OPEC supplies follow a similar pattern in BAU, expanding in the first half of the Outlook, led by increases in US tight oil and Brazil, before declining in the second half as US tight oil peaks in the early 2030s. This reduction provides scope for OPEC to increase its production from the mid-2030s onwards, with its level of output by 2050 back close to 2018 levels and its market share rising to over 40%.
As well the overall demand for liquid fuels falling, the transition to a lower carbon energy mix also prompts a shift in the composition of liquid fuels. In Rapid, the overall decline in liquids supply over the Outlook is more than accounted for by a sharp fall in crude and condensates, while the production of biofuels increases by over 2 Mb/d. Similarly, although total liquids supply by 2050 in BAU is only slightly lower than 2018 levels, this is more than accounted for by a larger fall in crude and condensates, partially offset by growing supplies of natural gas liquids (NGLs) and biofuels.
The carbon emissions associated with the production and transportation of crude oil and condensates accounted for around 5% of total carbon emissions from energy use in 2015.
There is a significant variation in these operational emissions – measured by the carbon intensity of crude supplies – across (and within) different countries, reflecting differences in the nature and location of the operations. These differences in carbon intensity affect the exposure of different types of production to carbon prices.
At low levels of carbon prices, these differences in carbon intensity have relatively little impact on overall costs and competitiveness and hence on the pattern of aggregate supplies. For example, in BAU in which carbon prices remain relatively low over the entire Outlook, the shift in the pattern of supplies between those in the highest and lowest quartiles of carbon intensity is less than 0.5 Mb/d by 2050 compared with a counterfactual case in which all supplies are assumed to have the same carbon intensity.
In contrast, the higher level of carbon prices in Rapid increases the additional cost levied on supplies with higher levels of carbon intensity and so has a more material impact on their competitiveness. In particular, in Rapid, the increasing carbon price causes highest quartile intensity supplies by 2050 to be around 2 Mb/d lower than in the counterfactual case (a decline of almost 25%), with correspondingly greater volumes of lower quartile intensity supplies.
The precise extent of this shift between high and low carbon intensity supplies depends on the extent the carbon intensity of different crudes and condensates can be reduced and at what cost. For example, it might be possible to reduce operational emissions of some onshore production via electrification at relatively little cost. But it seems likely that some of the differences in intensity is likely to persist and so have a bearing on the future pattern of supplies if carbon prices increase materially.
A similar set of issues applies to differences in the carbon intensity of natural gas supplies, which in addition to differences in the carbon intensity of production, also depend on whether the gas is transported via pipelines or as LNG, which adds to its carbon intensity.
The outlook for refining is downbeat, reflecting the impact of Covid-19 in the near term and a combination of declining liquid fuels demand and increasing competition from alternative feedstocks further out.
Refinery throughputs in both Rapid and BAU are significantly lower in the near-term as a result of Covid-19 reducing the demand for refined products, especially in the transport sector (see Oil).
As with the overall demand for liquid fuels, refinery runs in Rapid never fully recover to pre-Covid levels and fall by more than 45 Mb/d to less than half of their 2018 levels by 2050. The outlook for refining is a little less challenged in BAU, with refining runs recovering close to pre-Covid-19 levels over the next few years and remaining around these levels until the early 2030s, before gradually declining to around 10 Mb/d below 2018 levels by 2050.
This outlook for falls in refining throughput contrasts with previously announced plans to build roughly 9 Mb/d of new refining capacity over the next 5 years or so. The scale of excess refining capacity could be even greater if emerging economies in which product demand continues to increase – such as India and Africa – build additional capacity to limit any increase in import dependency for refined products.
The excess refining capacity that emerges in both BAU and Rapid leads to increasing competition and the eventual shutdown of the least competitive refineries. The shutdowns in BAU are concentrated in the developed economies – particularly Europe, OECD Asia and parts of North America – where falling domestic demand increases refineries’ exposure to the highly competitive product export market.
The degree of market rationalization required in Rapid is more pronounced and widespread, with around 50 Mb/d of current (or planned) capacity surplus to requirements by 2050. The refining capacity that is most resilient to these pressures in Rapid is aided by: resilient domestic demand; access to advantaged feedstock, high levels of upgrading, integration with petrochemicals and, in some regions, government support.