December 31, 2021

Decarbonization Supply Curve for the United States

There is no current, published greenhouse gas abatement supply curve for the United States [as of January 2022].

Scores of supply curves (aka marginal cost curves) have been published with a global scope (notably, Goldman Sachs’ 2020 “Carbonomics” report, and McKinsey’s 2010 version). And some supply curves for individual U.S. states are available (notably, reports using E3’s Energy Pathways software). But, with the United States (or US + Canada) as the geographic scope, there is nothing out there–not since McKinsey’s 2007 curve that, today in 2022, is uselessly outdated in terms of the list of commercially viable technical solutions, abatement costs, and projected BAU emissions levels by source (given decarbonization action to date and policy changes already in place).

So, I made one.

The tool below builds a supply curve by rank-ordering technical solutions for greenhouse gas abatement by cost. (“Cost” = levelized net cost to society, presented in annual US dollars per ton of carbon dioxide equivalent tonnage—where cost means incremental capital and operating costs less any fuel cost savings, without any top-line revenue from a market for monetizing such abatement actions).

The x-axis represents the volume of abatement potential from each solution. (“Volume” = annual abatement CO2e tonnage in 2030).

Please enter your own cost and quantity assumptions for each technical solution below to produce your own curve. (Make sure to read further below for important guidance about properly using this tool, and to review the decarbonization context in the United States):

GHG abatement Chart
Technical solution for GHG abatement  Net cost to society (levelized $/ton CO2e)  Annual CO2e abated (millions of tons, in 2030)
A Improved thermal efficiency of building envelopes and energy efficiency of building operations (improvements beyond current BAU projections)
B Improved cropland management and livestock management
C Electrification of road transportation - PASSENGER CARS AND BUSES, SHORT-HAUL FREIGHT (also includes passenger ferries)
D Improved forest management
E Industrial material and energy efficiency improvements, process shifts, and electrification (waste heat recovery, pump sizing, recycling consumer wastes, switch to biogas fuel) Fugitive high-GWP gas mgmt shown separately.
F Control of high-GWP fugitive gases (methane, F-gases)
G Fuel substitution in road transportation -LONG-HAUL FREIGHT (liquid biofuel replacing diesel, H2 fuel cells)
H Zero-carbon power generation - NEW UTILITY-SCALE PV (with enabling investments in T&D and energy storage via batteries/H2/pumped hydro)
I Zero-carbon power generation - NEW WIND (with enabling investments in T&D and energy storage via batteries/H2/pumped hydro)
J Zero-carbon power generation - NEW RENEWABLES OTHER THAN WIND AND UTILITY-SCALE PV (distributed solar, biomethane combustion with CCS, run-of-river hydro, solar thermal -- with enabling investments in T&D and energy storage via batteries/H2/pumped hydro)
K Afforestation and reforestation
L Direct electrification of building operations - NEW BUILDINGS (space and water heating, cooking)
M CCS from flue gasses of new and existing industrial plants and existing fossil fuel power plants
N Zero-carbon power generation - SMALL MODULAR NUCLEAR REACTORS
O Aviation fuel-switching to liquid biofuel
P Fuel substitution for maritime shipping (ammonia fuel, liquid biofuel replacing diesel, liquefied biomethane replacing LNG)
Q DAC and storage of atmospheric carbon
R Green hydrogen or biogas fuel/feedstock in industrial processes (especially primary steelmaking)
S Electrification of very short-haul aviation (batteries or hybrid H2-electric)
T Direct electrification of building operations - EXISTING BUILDINGS (space and water heating, cooking)
U Something else: _________________________

Decarbonization context in the United States

  • The United States produces about 10% of global CO2e emissions (5.2 billion out of the 50 billion annual tons, in 2020) while representing 25% of global GDP. The U.S. is thus disproportionately MORE efficient than the rest of the world—producing more economic output with relatively less emissions consequence. At the same time, however, the U.S. is home to only 4% of the global population; thus, on a per capita basis, the U.S. contributes disproportionately to global emissions—which is relevant from the sense of fairness to individuals.
  • In the United States, as of 2022, greenhouse gases other than CO2 (i..e., methane, nitrous oxide, and F-gases) represent approximately 15% of total carbon dioxide equivalent tonnage (CO2e).
  • CO2e emissions in the U.S. are currently produced by industry and agriculture/livestock activities (30%), non-electric transportation (30%), electric power production (30%), and operation of commercial and residential buildings (10%). 
  • Direct air capture (DAC) has theoretically infinite volume potential as an abatement solution. (However, some would argue that it’s unclear that DAC could scale to such a volume.) At this time, DAC effectively functions as the ceiling on the marginal cost of meeting most GHG abatement goals.
  • Abatement solutions will be implemented sequentially, beginning with the easiest (where “easy” means controlled by a central decision point and technically feasible at affordable cost):
    • Phase I – Already decided and proceeding on a momentum path. (Much of this is included in current BAU GHG emission projections.)
      • New, zero-carbon electric power production (with accompanying investments in requisite T&D[1] and energy storage[2])
      • Thermal and energy efficiency improvements in existing and new building envelopes and operations
    • Phase 2 – Today’s current focus of tactical execution plans
      • Direct electrification of passenger transportation and short-haul freight (NB: Some of this is included in current BAU GHG emission projections.)
      • Control fugitive high-GWP gases
    • Phase 3 – Today’s current focus of feasibility studies and pilot projects
      • Changes to industrial processes (switch to biogas fuel, process shifts, energy efficiency and material efficiency improvements)
      • Carbon capture and storage (CCS) from flue gases of existing and new industrial plants, existing fossil-fuel power plants
      • Improvements in agricultural/livestock practices
      • Liquid biofuel to replace diesel for a portion of long-haul trucking, maritime shipping, and aviation
      • Direct electrification for water-heating and space-heating of buildings
      • Switch legacy methane-fueled electric power production to biomethane plus CCS
      • Reforestation, afforestation, and improved forest management
    • Phase 4 – Not yet commercially viable (i.e., technically and/or financially) [* not included on this curve]
      • *Green hydrogen fuel for electric power production (liquid H2 combustion turbines, or H2 fuel cells)
      • Green hydrogen fuel cells for long-haul trucking
      • *Green hydrogen fuel for water/space heating of buildings (H2 gas pipelines)
      • *Green hydrogen fuel for aviation (liquid H2 combustion, or H2 fuel cells)
      • Green hydrogen fuel (or interim biomethane) for industrial processes (e.g., H2 totally replaces NG as the chemical reductant in DR-EAF steelmaking[3]; H2 partially replaces coal coke as the chemical reductant and fuel in BF-BOF steelmaking blast furnaces[4])
      • Green ammonia fuel for maritime shipping[5]
      • *Green ammonia replaces conventional ammonia feedstock in fertilizer production
      • Electric batteries for very short-haul aviation
      • New small modular nuclear reactors for electric power production
      • Direct air capture (DAC) and storage of CO2

Important guidance on using this tool

You cannot abate more tonnage than what is emitted. This logic applies to projected 2030 U.S. BAU emissions of 4550 million tons per year, and it applies to BAU emissions from individual sources. Error messages will pop up if you enter abatement volumes that exceed BAU volumes.  

This tool bounds the universe of GHG abatement solutions, as follows:

  1. Some Phase 4 solutions are NOT on this curve—particularly those technologies with a long path to technical viability (e.g., nuclear fusion, turquoise hydrogen[6]).
  2. Only “technical” abatement actions are included. This curve does NOT include human behavioral changes that reduce demand for the carbon-emitting activity (e.g., reduced meat consumption per capita, slower shipping speeds, improvements to long-haul trucking fuel economy via driver behavior, changing home thermostat setpoints, or carbon labelling to enable consumer decisions to push supply-chain decarbonization).
  3. Technical abatement solutions that preserve the status quo are NOT included:
    • Coal-to-NG power plant conversions. Results in a net INCREASE in carbon emissions over time, given plant lives.
    • Relicensing existing nuclear power plants
    • Biomethane to replace natural gas fuel in new or existing commercial and residential buildings (for water-heating, space-heating, cooking). Can’t capture the CO2 and high-GWP emissions from dispersed combustion sites. Moreover, electrification is cheaper in new builds, gas combustion creates indoor air quality issues, consumer NG prices are more volatile than electricity prices and biogas would likely be similar.
    • Biofuel applications in transportation have lost favor for situations where electrification is possible (biomethane to replace natural gas in passenger vehicles and buses; liquid biofuel to replace diesel in passenger transport and short-haul trucking). Biofuels are considered carbon-neutral, but are only carbon-negative with accompanying CCS (which is not possible for mobile, dispersed emissions sources).
    • Blue hydrogen (i.e., reforming and splitting methane into hydrogen and carbon dioxide + CCS of the carbon dioxide)
  4. Technical abatement solutions that are already part of the BAU path NOT included:
    • Better fuel efficiency (miles per gallon) standards for combustion engine vehicles and aircraft. Better energy efficiency (kilometers / kWh) for electric vehicles.
    • Demand-side management programs to reduce electricity consumption behavior of consumers
    • Coal plant closures. A foregone conclusion in the U.S., as they all are—or soon will be–cheaper to close down than to operate.
    • NB: Some portion of the following GHG abatement solutions are already included in current BAU GHG emissions projections: new renewable electric power generation, further thermal and energy efficiency improvements in existing and new building envelopes and operations, direct electrification of passenger transportation and short-haul freight.
  5. Technical abatement solutions that are not relevant to the United States are NOT included:
    • New, large-scale, third- or fourth-generation nuclear power plants. Unrealistic from a policy and cost perspective that this will happen any time soon in the U.S.
    • New, large-scale hydroelectric power plants. Not realistic in the U.S.  Could be feasible in Quebec, and thus relevant to a curve that includes Canada. But some research questions the net carbon impact, given methane released from reservoirs.
    • Biomass fuel or co-firing for power generation
    • Solutions to improve water usage efficiency and wastewater recovery. This only has a meaningful GHG abatement effect in locations where there are desalination operations to avoid.
  6. Enabling technologies that are required as part of other technical abatement solutions (but are not standalone solutions) are NOT included:
    • Electricity storage
    • Electricity load-shifting (e.g., demand-response programs)
  7. In order to use this tool to support YOUR decision analysis, you must additionally layer on your specific parameters and boundaries. An economic cost curve is very different from a list of available investment opportunities. The list of solutions and rank order of solutions is different when taking an economy-wide view vs considering the specific operating boundaries of a GHG emitter.
    • Eliminate solutions that you do not have control over or cannot invest in as a third party.
    • Consider the timing feasibility of solutions. Some solutions on this curve are expected to provide abatement potential by 2030 but are not commercially viable as of 2022.
    • Update the Y-axis values for each solution, to represent cost/ROI to the decision-maker (rather than cost to society, as this curve depicts). Consider the cost to the entity taking the action: capital cost, operating cost, and timing of those expenditures. Consider the revenue and/or cost savings that accrue to the entity taking the action.

Additional Q&A

  • “What about the Tonga volcano?”  Volcanic ash from a major eruption may indeed provide limited, temporary cooling for a few years. However, all that does is extend the date by which we need to hit those GHG abatement targets. We are already slipping against those goals, so this should not be cited as an excuse to slip more.
  • “What about the sequencing of abatement actions?”  The temporal aspect of ordering of actions does indeed matter and affect decarbonization results. For example, if we electrify lots of things before electric power production is low-enough carbon, we will increase GHG emissions along the pathway to long-term abatement.  However, we in the United States live in a market economy (and a federalist, non-unitary government system. . . and moreover one too divided on partisan lines to realistically take bold action) without central planning, and so we cannot realistically specify sequencing of actions by private corporations or mandate actions across state and local governments. Therefore, most participants in the decarbonization world do not worry about this, and I do not contemplate it as part of this supply curve tool.
  • What is the U.S. GHG emissions goal? Greenhouse gas emissions peaked in the U.S. in 2007 and emissions have been declining ever since. As of 2022, the Biden Administration’s goal is a 50% reduction of 2005 emissions (which were 6600 million tons of CO2e) by 2030. That means a target of annual emissions in the year 2030 of 3300 million tons; and, given currently projected BAU emissions in 2030 of some 4600 million tons, meeting the 50% target requires incremental annual abatement of 1300 million tons by 2030. As you can see on this GHG abatement supply curve (using the default pre-filled values for costs and volumes), the marginal cost of 1300 million tons of annual abatement is $100 per ton.
  • What do we mean by “biofuels”? Terminology varies in this space. I use “biofuels” to collectively refer to biogas and liquid biofuels:
    • Biogas = Captured from landfills, livestock ops, wastewater treatment; or produced from organic waste, dedicated energy crops, algae. Can be purified into biomethane/ renewable natural gas. Can be compressed or liquefied.
    • Liquid biofuel = Produced from dedicated energy crops, waste plant oils, waste animal fats, microbial lipids. These exist in liquid form without compression.

– December 2021

[1] Needed to support estimated 2.0-2.5x growth in electricity demand by 2050 under an “electrify everything” approach: widespread T&D upgrades, new long-haul DC regional interconnection for seasonal load balancing across regions, digitalization/smartgrid controls, and voltage regulation tech.

[2] Achieving high penetration of non-dispatchable / intermittent power generation resources requires production-smoothing, utility-scale storage technologies: lithium-ion batteries, pumped hydro storage, flow batteries and other emerging battery technologies, compressed air, flywheels, emerging molten salt technology.

[3] About two-thirds of U.S. primary steelmaking employs the DR-EAF process. H2 would totally replace NG as the chemical reductant (reducing solid iron ore to solid elemental iron without melting it). The electric arc furnace would operate as usual to melt the iron and produce molten steel.

[4] About one-third of U.S. primary steelmaking employs the BF-BOF process. H2 would partially replace coal coke as both the chemical reductant and the fuel in the blast furnace (which reduces solid iron ore to liquid elemental iron). The basic oxygen “furnace” (autogenous – no fuel needed) would operate as usual to convert molten iron to molten steel.

[5] Green ammonia = Hydrogen from renewable-electricity-fueled water electrolysis + Nitrogen from renewable-electricity-fueled air distillation. Renewable-electricity-fueled Haber-Bosch process to combine the hydrogen and nitrogen into NH3. (Conventional ammonia = Hydrogen from steam-reformed methane + Nitrogen from fractional-distilled air. Haber-Bosch process to combine the hydrogen and nitrogen into NH3.)

[6] Turquoise hydrogen = methane pyrolysis into hydrogen and solid carbon