[Updated with version 1.1 September 19, 2024]
Click the link below to download the LNG Spreadsheet Tool (LNGST, version 1.1). The model was developed by Symons Public Affairs to help improve the state of knowledge of LNG's GHG and economic impacts, as well as the assumptions underlying LNG analysis. LNGST v1.0 was developed based on input from reviewers of an earlier review draft.
This post includes a discussion of key factors to consider in using LNGST. To maximize user control, LNGST documents and explains recommended ranges of input assumptions. It includes a user interface to adjust input assumptions, even when those assumptions are outside recommended ranges.
About LNGST
The LNG Spreadsheet Tool is designed to assess a given volume of LNG that is being considered for approval by DOE. The tool includes an updated list of LNG permit applications and volumes. The starting setting is based on the volume of LNG -- 12.3 billion cubic feet per day (Bcf/d) -- of the eight LNG projects that are currently in late- and middle-stage permitting. The approved volume of these eight projects, which include CP2 (the largest US export terminal ever proposed), is equivalent to 90% of all residential natural gas consumption of the United States. These are also the projects most impacted by the proposed Energy Permitting Reform Act.
For outputs, LNGST calculates:
Annual and cumulative greenhouse gas emissions using 20-year and 100-year GWPs
Coal-fired power plant emission comparisons
Increased US energy expenditures
Avoided climate damages, based on the US EPA's Social Cost of GHG calculator
The amount of renewable energy electricity transmission that would be needed to equalize the LNG emissions.
Background
Gas export decisions are a critical component of US energy and climate policy. Dr. Robert Howarth of Cornell University has completed a study that has been peer-reviewed, updated, and accepted for publication in a peer-reviewed journal. This peer-reviewed study concludes (emphasis added):
"Overall, the greenhouse gas footprint for LNG as a fuel source is 33% greater than for coal when analyzed using GWP20, (160 g CO2-eqivalent/MJ vs 120 g CO2-eqivalent/MJ). Even considered on the time frame of 100 years after emission (GWP100), which severely understates the climatic damage of methane, the LNG footprint equals or exceeds that of coal."
The Department of Energy (DOE) is currently updating its analysis of the greenhouse gas (GHG) emissions of US LNG exports. Under the Natural Gas Act, DOE has the responsibility to evaluate whether authorizations for the export of LNG to non-free trade agreement (NFTA) countries is consistent with the “public interest." Eight projects, with pending applications at DOE are in the late- or mid-stages of permitting (see project update here). These projects total 12.3 billion cubic feet per day (Bcf/d) of volume, which is equivalent to 90% of all residential natural gas consumption of the United States. In order to attract the $75 billion in investment needed for these projects, developers rely on DOE approval to export LNG to China and other nations without free trade agreements.
According to DOE Deputy Secretary David Turk (testifying before Congress):
"Since 2018, there have been truly transformative changes that need to be fully incorporated in our analysis. And, in turn, this updated analysis will allow us to better address a wide variety of key questions that need to be answered for us to make public interest determinations. First, the amount of U.S. natural gas that is being exported has dramatically increased, and we need to answer how authorizing exports beyond these unprecedented volumes could impact affordability for U.S. consumers and competitiveness of U.S. manufacturing. Second, our understanding of CO2 and methane’s effect on climate change have only become sharper, and we need to further improve our analytical tools to answer a range of questions about LNG exports’ climate and environmental consequences, both near and longer term. Third, increased deployment of clean energy is driving updated estimates of fossil fuel demand and usage over time; and we need to understand how the latest regional and global trends will impact our own energy security as well as that of our allies."
I and other observers have been critical of DOE's 2019 GHG study, which was issued at a time when the United States had publicly repudiated its global climate commitments (see my report here and memo here). As indicated by Deputy Secretary Turk, the 2019 study is out of step not only with the latest climate science, but also with the United States' domestic and global climate goals. Most notably, the study compared LNG to other fossil fuels without considering the rapid transition to clean energy alternatives needed to meet the objectives of the Paris climate agreement, which calls for limiting climate change to well below 2°C, nor the Glasgow Climate Pact, which calls for a global phase down of fossil fuels.
The Council of Environmental Quality's 2023 GHG NEPA Guidance instructs agencies to evaluate "to what extent the proposal's reasonably foreseeable GHG emissions are consistent with GHG reduction goals, such as those reflected in the U.S. nationally determined contribution under the Paris Agreement." The guidance also directs the agency to quantify not only the net impacts, but also "the reasonably foreseeable gross GHG emissions increases." DOE should apply the same principles in analyzing LNG. The emissions from gas production, transmission, liquefaction, shipping, degasification, and combustion all result in measurable and attributable emissions to the atmosphere. These emissions use up a portion of the limited remaining GHG budget of allowable global emissions before we exceed target levels.
Time Horizon
The LNG Spreadsheet Tool is designed to assess new approvals of LNG beyond projects that are currently operating or under construction. After DOE approves an LNG application, it can take up to 7 years for the applicant to secure financing, construct the facility, and begin operations. DOE permits are granted for a 25-year period. LNG projects require enormous capital (billions of dollars) and are based on expected lifetimes of 30-50 years. The spreadsheet tool makes a simplifying assumption that newly approved projects come online in 2030 and measures impact through 2050.
The following sections provide an overview of some of the most critical factors when assessing the GHG impacts of LNG.
The Precautionary Principle vs. Wishful Thinking
There are different layers of uncertainty in assessing the GHG impacts of expanding US LNG exports. That uncertainty has been reduced considerably in recent years when it comes to calculating lifecycle emissions of LNG. Disagreements about how to calculate net impact on world energy markets, on the other hand, have only grown more pronounced as the world has moved away from incremental GHG reduction goals toward transformative, net-zero pathways that require the rapid phase down of fossil fuels.
In assessing uncertainty, the precautionary principle applies. The US has signed numerous treaties that recognize the precautionary principle, including the 1992 Rio Declaration and the Framework Convention on Climate Change, which was unanimously ratified by the Senate. The precautionary principle reads (as stated in the Rio Declaration):
"Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation.”
Applying this principle to evaluating LNG impacts, the public interest is served when steps are taken to increase the availability, use, and cost competitiveness of clean, renewable energy sources and energy conservation and to phase down fossil fuels. Uncertainty should never be used to hide the true potential for significant climate damages from decisions that increase long-lived fossil fuel infrastructure. And yet this precautionary approach to analyzing impacts is too often ignored.
As the LNG toolkit demonstrates, there are a number of modeling assumptions that have significant impacts on the results. For any one of these assumptions, there is a "wishful thinking" end of the range, where emissions impacts are minimized, and a worst-case scenario, where potential emissions are maximized. Simplistic and improbable assumptions, such as the oft-cited notion that LNG will displace global coal consumption rather than increase overall fossil fuel consumption or displace clean energy, are not only wrong but also a dangerous deception.
In particular, there is a danger of sliding all the uncertainty levers in the same direction to pile improbable assumption upon improbable assumption, to the point that investing billions of dollars in expanded gas exports is proclaimed to be as a net-positive for climate change. These cases are beyond wishful thinking, to the point of being the modeling equivalent of a Goldilocks scenario. They not only violate the precautionary principle, they violate basic science. We can't phase out fossil fuel emissions by building infrastructure intended to expand production, especially when the fossil fuel has an emissions footprint worse than coal.
Some have argued that permits don't matter because global LNG demand is projected to peak before 2030. This creates the risk of an analytical fallacy, where permits would be issued based on the assumption they won't be used. Models based on free market economic theory can be surprisingly blind to the billions of dollars in long-term LNG contracts that have already been signed to back these proposed projects. It also ignores the fact that LNG developers secure billions in financing based in large part on 20-year purchase contracts.
The precautionary principle should again apply here. If billions of dollars are backing LNG projects that are only economic if they produce LNG for decades, then the government must assume that the LNG will be used. Also, models ignore the fact that expanding exports to more global fossil fuel markets also exports the powerful fossil fuel lobby that fights to protect its markets from all threats, especially renewable energy.
The lessons learned from a similar energy deal in 2015 should be enough to give anyone pause when arguing that LNG will not be used if approved. The 2015 budget deal paired renewable energy tax credits with a provision to lift the decades-old ban on exporting US crude oil. Energy analysts rushed to validate the bargain. Those clean energy provisions would “dwarf the impact on carbon emissions of allowing oil exports,” wrote Michael Levi in an analysis widely quoted at the time.
To quell fears about the oil provisions, Nancy Pelosi sent a letter to fellow Democrats, writing: “While lifting the oil export ban remains atrocious policy, the wind and solar tax credits in the Omnibus will eliminate around ten times more carbon pollution than the exports of oil will add.” Her appeal worked. The bill passed. Contrary to the assurances of energy experts, the oil export floodgates opened. Crude exports surged from zero to four million barrels a day today. This growth in exports was 20 times higher than the worst-case scenario forecasted in 2015 by Levi, the U.S. Energy Information Administration (EIA), and others.
Methane
Natural gas is primarily methane. When emitted to the atmosphere, methane has an 82 times higher warming impact than when it is burned and converted to CO2. This is over a 20-year period, which is a more appropriate time horizon for confronting climate change than conventional 100-year time horizons (LNGST provides results based on both the 20-year and 100-year GWPs). As President Biden said to world leaders in Glasgow, this is the "decisive decade" to act.
Fortunately, the scientific basis for understanding the lifecycle greenhouse gas emissions associated with US LNG exports has increased significantly in recent years, reducing uncertainties and better capturing the full climate impact, especially from methane leaks upstream of LNG facilities. The lifecycle estimates of the directly attributable emissions entering the atmosphere due to fossil fuel production and consumption are understood with far greater certainty than trying to assess market effects.
The greatest source of uncertainty for calculating lifecycle (gross) emissions is methane leak rates in the United States from oil and gas production, processing, compression, and transmission. Numerous peer-reviewed studies, based on more than a million aerial and satellite measurements, have concluded that observed leak rates are three times as high as government estimates nationwide, and up to 10 times as high in the Permian Basin. As all pending LNG proposals would be based along the Gulf Coast, the Permian Basin (covering parts of Texas and New Mexico) would be the leading supplier of US natural gas. According to U.S. EIA's Annual Energy Outlook 2023, about three-quarters of increased gas production due to LNG would come from the Gulf Coast and Southwest regions.
Together, these findings of high Permian leak rates and the sourcing of LNG expansion from the Permian region upend prior assumptions that US upstream methane emissions for the gas supplying LNG expansion would be better than from other global sources. They also suggest that US LNG will have a larger greenhouse gas footprint than the two biggest LNG suppliers beyond the US: Qatar and Australia, which the International Energy Agency identifies as having methane leak rates that are similar to or lower than the US national average.
In addition to allowing users to adjust methane leak rates for the US and other nations, the spreadsheet tool includes a recommended methane leak of 2.8% (central point, range of 2.3%-4.0%) that is based on the 2024 peer-reviewed study by Sherwin (LBNL) et al, published in Nature. The study assessed nearly one million aerial measurements and concluded that nationwide average methane emissions are 2.95%, “roughly three times the national government inventory estimate." Emissions averaged 4.60% when excluding the Pennsylvania study, which is relevant for LNG given pipeline limitations to move gas from PA to the Gulf Coast. Emissions were especially high (up to 9.63%) in the Permian Basin. The methodology for the recommended range and central point are explained further in the spreadsheet tool, Tab B ("Discussion").
Some have argued that DOE should base approval decisions on promises to certify gas as having low methane leaks. There are several problems with this approach. Once DOE issues a permit, developers receive license to operate through 2050, with no oversight controls built in for DOE to assess certification. Second, a vast portion of oil and gas methane leak volume is from large, unexpected and undetected leaks. No pledge or program can accurately guarantee that a given source of gas supply will entirely avoid these methane leaks. Not surprisingly, independent observations have demonstrated that certified gas products have had multiple undetected methane leaks. Finally, putting a certification label on any given stream of gas will not stop production from increasing in fields with high methane leaks. A gas certification scheme cannot even meet the standards set in DOE's “Voluntary Carbon Markets Joint Policy Statement and Principles," which states that claimed emission reductions must “prevent emissions from occurring, being shifted, or intensifying beyond their boundaries as a result of the activity."
The "Net Emissions" Paradox
Compared to well-understood "gross" LNG lifecycle emissions, "net" emissions assessments of the impacts of expanded LNG on global and US energy flows and emissions are largely guesswork. Traditional energy models, which assess market responses to changes in supply and demand, shed little light when it comes to the central GHG question DOE should ask when determining the public interest, which is: "Does the LNG move us closer or further away from achieving our climate goals?"
In order to measure whether approving additional LNG capacity is consistent with global climate goals, the appropriate baseline for comparison should be a deep decarbonization scenario that achieves net-zero global emissions by 2050. To assess this question, DOE must compare LNG volumes to a deep decarbonization pathway, such as the International Energy Agency's World Energy Outlook 2023 Net Zero Scenario.
This creates a paradox for traditional energy modeling, which assesses change against business-as-usual scenarios. In a net-zero pathway world, there is no demand for natural gas from these facilities and the facilities should not be built. If they are built, they won't be used. Either assumption creates a paradox since the entire purpose of applying for a permit is to build and use the facility.
Modelers avoid this trap by making assumptions based on historic supply/demand responses and applying them to a business-as-usual baseline. That is, they are no longer answering the question "does this move us closer or further away from a deep decarbonization scenario?" Rather, they are asking "does this change anything from a business-as-usual scenario, which is typically a case that results in a 3 degree or higher temperature change?" Under these conditions, a result of "zero" net emissions only tells us that the LNG is locking in and affirming the three degree (or more) pathway.
A better approach for DOE would be to compare LNG emissions to energy benchmarks for a deep decarbonization scenario, rather than cause-and-effect supply/demand models. Given that U.S. and global climate goals require a net-zero decarbonization pathway, DOE should assume that these pathways are in the public interest. The relevant question is to what degree long-term LNG approvals deviate from this pathway.
LNGST seeks to provide recommended inputs, but also the ability to insert alternative assumptions. Consequently, I included a "3 degrees plus" net calculation to show how traditional market leakage studies might be used to assess substitution, but also to clarify that this does not answer the public interest question before DOE. Rather, it measures change from a BAU scenario that will lock in climate change of 3 degrees or more.
For the "Under 2 degrees" scenario, which is what DOE should use as the basis for public interest determination, I recommend using the energy mix from IEA's WEO 2023 net-zero pathway, which is shown and calculated in one of the worksheets.
Coal
Much ado has been made about LNG purportedly displacing coal, but this claim is not backed by historical evidence, and it certainly falls flat when looking at current emissions trends and deep decarbonization scenarios.
China is the logical place to examine. It is the largest importer of LNG in the world. Two Chinese banks (the Bank of China and the Industrial and Commercial Bank of China, or ICBC) have helped finance eighty percent of US LNG projects that are operating today or have fully secured financing. China is among the biggest financial backers of proposed LNG projects currently pending at DOE, based on long-term purchase commitments. Venture Global, the company behind the massive CP2 LNG project in Louisiana, has contracts to sell 11.8 million tonnes of US LNG annually to Chinese firms (Sinpoec, China Gas, and China National Offshore Oil Corp) across its US LNG projects.
And yet, there is no evidence that LNG has displaced coal globally or in China. In June 2024, the Institute for Energy Economics and Financial Analysis (IEEFA), issued a paper titled "LNG is not displacing coal in China’s transition to renewable power." According to the report, the share of natural gas generation in China's power sector (which accounts for 60% of China’s total coal usage) has remained at just 3% since 2015 despite big upticks in LNG. Meanwhile, the share of wind and solar generation has quadrupled to 16%.
Looking forward, the economics are even less favorable for gas replacing coal in China. In all scenarios (including Stated Policies and Net-Zero) of the 2023 World Energy Outlook, IEA finds that the capital costs of PVs in China will fall below the capital costs of gas (and coal) before 2030 (Table B4). In the Stated Policies scenario, which has the least deployment of clean energy, IEA (Table A.1a) projects coal consumption to peak in every region of the world by 2030 and decline significantly thereafter. IEA (Table A15) projects that China coal consumption will decline 13% between 2022 and 2030, and 45% between 2030 and 2050, in this Stated Policies scenario. In the Net-Zero scenario, global coal consumption is projected to decline 98% by 2050.
Coal functionality is nevertheless included in LNGST so users can assign their own values.
US Natural Gas Consumption
US EIA is warning that currently low Henry Hub gas prices are temporary because LNG exports will once again drive up energy prices. From the September Short Term Energy Outlook:
"We forecast natural gas prices will remain relatively flat in the upcoming shoulder season during September and October before generally rising in 2025. Price increases in 2025 reflect U.S. natural gas production that does not keep pace with growth in U.S. liquefied natural gas (LNG) exports. We expect the Henry Hub spot price will rise from less than $2.00 per million British thermal units (MMBtu) in August to around $3.10/MMBtu next year."
As far back as 2015, EIA has warned that higher overseas demand for U.S. LNG exports raises the average Henry Hub price ... 35% above the Reference case price." They were right. American consumers spent $142 billion more on natural gas in 2022 than 2016. Households have seen their gas bills go up more than 50% since 2015, more than double the rate of inflation. This occurred despite a 45 percent increase in gas production between 2016 (the year that the US began exports) and 2022.
Increased prices can reduce domestic (US) natural gas consumption. The spreadsheet tool includes recommended values that are based on comparisons of LNG cases from U.S EIA's 2023 Annual Energy Outlook.
It's important to note that what is true for LNG analysis must also be true for analysis of other sectors if comparisons are going to be made between models. That is, if restricting US gas supply (through exports) increases supply and decreases US gas consumption, then it also holds true that an increase in renewable energy sources will free up gas supply compared to baseline scenarios, which should decrease gas prices and increase US gas consumption as well as emissions in other sectors.