Oxy Low Carbon Ventures & The DAC Landscape

takeaways from oxy’s low carbon ventures investor day & the dac landscape today

*Post was originally written on April 4, 2022. You can find a related twitter thread summary here.

Oxy hosted their Low Carbon Ventures day on March 23 and I wanted to highlight some of it given it was one of the first examples we’ve seen of a public company outlining the long-term opportunity related to DAC – scale, economics, business model, funding plan, etc. Materials from their investor day can be found here.

Cost expectations: The Levelized Cost of Capture (LCOC) from the first DAC plant (DAC1) is expected to be $300-425/ton, declining to $250-350/t in 2025-30, and ultimately reaching <$150/t longer-term. Capital costs come from Air Contactors (40%), Centralized Processing (40%), and the rest from Utilities and Infrastructure (20%). Oxy expects 15-20% reduction in LCOC from scaling their plants and see the main areas of efficiency coming from moving air, growing pellets, and process intensification, among other categories.

The biggest hypothesis will be whether Oxy can scale down costs as quickly as they think. The ‘end-state’ vision that many DAC startups and companies call out is <$100/ton LCOC but we have a very long way to go in order to achieve that, especially as these plants represent the first of their kind in terms of the technology at scale (more on that below).

Economic Model: Oxy expects revenue to come primarily from carbon removal credits, policy support, and higher priced oil (relative to peers). Based on their estimates, they think up to ~5 gigatons (billion tons CO2e) of emissions from the hardest-to-abate industries would be economical at $250/ton per credit. They also only plan to only build the plants in areas where policy is support. In the US, the 45Q tax credit is $35-$50/ton and you can also stack it with other tax credits (e.g., State-specific example of CA LCFS which has been trading >$100/ton). They also expect to sell premium priced ‘Net Zero Oil’ and have already struck one deal with SK trading which gave the South Korean refiner an option of buying up to 200,000 barrels a year.

The core part of the economic model from my view is whether they will be able to sell credits at $250/ton and if there will be a deep enough inventory of buyers to buy removal credits at those levels. The voluntary carbon markets (VCMs) are still in their nascency all things considered but currently, there aren’t many mechanisms from preventing buyers from buying the cheapest possible credits.

Pulling from a publicly available dataset (given the fragmented nature of VCMs, one must assume this is a partial snapshot), the weighted average removal price was ~$8/ton in both 2021 and 2020. One can reasonably assume the standards and rigor around quality of removal purchases will improve over time pushing corporate buyers towards higher quality credits (permanence, additionality, verification, etc.) which should push up the price of more scarce, high quality removal credits. However, it shows the current gap between the state of markets and where they need to go.

If you look at lists of durable (100+ years) carbon dioxide removal (CDR) purchases, you’ll see higher prices. This dataset reports an average of $332/ton. However, many of these purchases are proof-of-concept purchases to help prove out the technology/processes vs. at-scale volumes. This is evidenced by the fact that the market is very small: ~69,349 tons sold, 30 CDR sellers, 12 buyers.

This further emphasizes the importance of the Oxy-Airbus deal.

• 400,000 tons is the largest nominal purchase of durable CDR to date by far (~5.8x all purchases to date based on purchase data per the above database)
• 100K tons per year represents ~20% of the plant’s expected initial capacity (500K tons per annum)
• At a price of $250/credit, that would represent a $100M purchase price
  • To be clear, there is no indication Airbus purchased the credits at this level but used to illustrate the size of the deal at the price Oxy wants to underwrite these investments at

Scale: OXY plans to have 70 DAC facilities with 1 MTPA capture potential each by 2035 (~1.5% of addressable market) with the potential for up to 135 plants (~2.5% of addressable market). They use the aforementioned 5,000 gigatons segment as their refence addressable market in this case.

In IEA’s recent DAC report, it talks about the immense scale-up we need in DAC but also gives a frame of reference to evaluate how ambitious of a plan this would be from Oxy.

• IEA Direct Air Capture Report
  • “In the IEA Net Zero Emissions by 2050 Scenario, DAC technologies capture more than 85 Mt of CO2 in 2030 and around 980 MtCO2 in 2050, requiring a large and accelerated scale-up from almost 0.01 MtCO2 today.” MtCO2 = million tons CO2.
  • “Currently 18 DAC facilities are operating in Canada, Europe and the United States. All but two of these facilities sell their CO2 for use, and the largest such plant – commissioned in Iceland in September 2021 – is capturing 4,000 tCO2/year for storage (via mineralisation).”
    • “The first large-scale DAC plant of up to 1 MtCO2/year is in advanced development and is expected to be operating in the United States by the mid-2020s.” – (Oxy’s plant).
  • “Future capture cost estimates for DAC are wide-ranging and uncertain, reflecting the early stage of technology development, but are estimated at between USD 125 and USD 335 per tonne of CO2 for a large-scale plant built today.”
    • “In locations with high renewable energy potential and using best available technologies for electricity and heat generation, DAC costs could fall below USD 100/tCO2 by 2030.”
  • “The scale-up of DAC deployment in the Net Zero Scenario implies an average of more than 30 DAC plants of 1 Mt/year being added each year during 2020-2050”
• Takeaways:
  • Bringing online this plant at the expected capacity of 500,000 tCO2 would already expand current global capacity ~50x, with current global operating capacity at <10,000 tCO2
  • Oxy then wants to build 69 more plants within the next 15 years (~5 plants/year)
    • using the IEA’s numbers to achieve a Net Zero pathway, Oxy would only represent ~17% of the required annual additions
    • A topic for a separate post, but these plants require significant energy, land, and heat which all introduce their own respective issues to scaling up this rapidly
    • Given each plant would take at least a few years, that means the buildout for 70 plants by 2035 requires serious acceleration to achieve or will be heavily dependent on productivity gains learned from building the first few plants

Funding:  DAC1 is expected to be $.8-$1.0 billion for the first 500,000 ton per annum train with a construction build-time of less than 3 years and an operating life of 25 years. The plant is expected to commence operations in 2H24. Through 2024, projects will be funded primarily through internal capital, government grants/loans, and strategic equity. From 2025-2030, the funding will transition towards financing through project debt and equity as opposed to deploying internal capital.

Overall, it will be really interesting to see how these next few years play out for Oxy and whether they can get any momentum behind this DAC strategy. This is one of the first examples we’ve seen of a public company investing heavily into a low-carbon business opportunity required for Net Zero pathways but represents an opportunity with significant technical and business model risk. I’ll surely be watching closely to see how their ambitions progress.