The Green New Deal: The enormous opportunity in shooting for the moon.

Decarbonizing with massive electrification will bring about a new American abundance.

Saul Griffith
Feb 21 · 15 min read

The conversation about a Green New Deal is bold, timely, and necessary. The most important component of the Green New Deal as currently drafted is its commitment to complete decarbonization. That is the destination, the moon for our moonshot. The question is: how do we get there? What we learn is it is at least twice as easy as we think.

The most important component of the Green New Deal as currently drafted is its commitment to complete decarbonization.

Traditionally, the discussion of the energy economy has been about efficiency and domestic production, and not about decarbonization. This is an artifact of the oil crises of the 1970s and concern for US energy independence. This old fashioned approach translated essentially to more fuel efficient (but still petroleum) cars and better insulated (but still natural gas heated) homes. This approach can get us only a small fraction of the way to total decarbonization. We need substitution technologies, ways of satisfying our energy needs with zero emissions, in addition to improving efficiency. Electricity generated from renewables (solar, wind, hydroelectricity, geothermal) and nuclear are the substitution technologies that meet the scale of the problem. The good news is that the path to total decarbonization using these technologies is clearer (and more compelling economically) than most people think.

Decarbonization is not an unattainable ideal; it follows directly from the best data available on US energy usage. In 2017, Otherlab, the company I founded and lead, was contracted by the Advanced Research Project Agency of the Department of Energy (ARPA-e) to review all available energy data sources and create an ultra-high resolution picture of the US energy economy. The purpose was to identify research priorities, and to model scenarios for new energy technologies and policies. This work leveraged many decades of effort by the EIA¹ and Lawrence Livermore National Lab² analyzing the US energy economy and providing annual snapshots in a Sankey³ Flow Diagram format. The Otherlab “Super Sankey” tool is available at www.departmentof.energy — yes, we were cheeky in choosing that URL.

The 1973 publication “Understanding the National Energy Dilemma”was the first use of Sankey diagrams for energy policy planning.

This work on mapping US energy usage began in 1973, when Jack Bridges, working for the Joint Committee on Atomic Energy (an intellectual precursor to the Department of Energy), used the Sankey format and improved upon it in a fabulous book, Understanding the National Energy Dilemma. The Committee had been charged to “develop an energy display system which, in less than an hour, could give an extremely busy person an understanding of the size and the complexity of our national energy dilemma.” That “busy person” was meant to represent the decision makers in Congress.

2018 “Super Sankey”. The diagram above is too small to read in detail; for best understanding, it requires a large wall display. The interactive online version at http://energyliteracy.com allows the viewer to drill down into any particular industry sector and understand the energy that it produces or consumes.

The width of a given flow identifies the size of a given energy use, and thus points the way to the most impactful interventions to reduce energy use or enable decarbonization.

The interactive Sankey diagram we created for ARPA-e in 2018 represents all of the energy flows in the US economy, from “primary energy” sources like coal, natural gas, oil, hydropower, solar, geothermal, and nuclear power, through their transformation into intermediate sources like gasoline, diesel, and other intermediate fossil fuels, or electricity, and then their use by various sectors of the economy. The diagram also represents the energy that is used in those transformations, including the energy used in generation, lost in transmission, or embodied in products and materials that use energy in their production. The width of a given flow identifies the size of a given energy use, and thus points the way to the most impactful interventions to reduce energy use or enable decarbonization.

Energy use at the scale of the national economy is measured in quads. One quad is 1 quadrillion British Thermal Units (BTU’s)³. Conveniently, the US uses about 100 quads each year, meaning that one quad is roughly equal to 1% of the US’s total national annual energy use.

What The Sankey Diagram Tells Us

Armed with the best data that we have, and charged with decarbonizing the economy, what type of picture, or set of options, emerges? The answer is so unexpected and so important that it should be central to any discussion of the Green New Deal.

The mere act of electrification has a powerful effect on efficiency: electrifying the US economy from carbon free sources will reduce the amount of energy needed by more than half.

Most experts agree⁴⁵ that the most probable and realistic pathway to massive decarbonization is massive electrification. The mere act of electrification has a powerful effect on efficiency: electrifying the US economy from carbon free sources will reduce the amount of energy needed by more than half. How can that be so?

Decarbonizing electricity saves us 25% of all national energy usage.

First, the largest saving in a totally decarbonized economy is the enormous amount of energy lost (and carbon dioxide produced) in generating electricity from fossil fuels. Around 25%⁶ of our total energy need is eliminated because burning fossil fuels to create electricity is a very inefficient process that generates an enormous amount of waste heat⁷⁸. Of the 39 quads of primary energy going into the electricity sector, only about 12 are actually delivered to industry, our homes, and our businesses. The rest disappears, quite literally, into the air as waste heat and carbon dioxide.

As we develop new primary sources of electricity, rather than converting fossil fuels to electricity, this waste is no longer needed.

Electrifying transportation saves us 14%.

The electrification of transportation is the next big energy win. 28 quads of primary energy goes into the transportation industry, which includes cars, trucks, airplanes, trains and boats. The overwhelming majority of vehicles now run on oil. If we electrify all cars and trucks, we will reduce by about ⅔ the amount of energy consumed in moving those vehicles⁹. Car engines are even less efficient than power plants in converting fossil fuels into a useful activity. Typically they run at about 25% efficiency, with further losses in the drivetrain. Electric cars have gone mainstream, are dropping in cost, and are expanding in performance, range, and options. A Green New Deal will increase the speed and impact of this industry transformation.

That’s great news; of the 21 quads we currently require for road and highway transportation, we’ll only need about 7, so the other 14 quads won’t need to be produced or generated.

Not needing to find, mine, refine and transport fossil fuels saves us 6%.

A huge amount of fossil fuel is used to discover, mine, process, and transport fossil fuels. In a zero carbon economy, we won’t need to expend that energy. How much is that? 0.9 quads are used for pumping natural gas through 4.4 million miles of pipelines; more than 3 quads is used to turn oil into gasoline and other fuel products in refineries; almost 2 quads is used extracting oil and natural gas; 0.25 quads is used for mining equipment and 0.25 quads is using diesel to push the coal from mine to power plant. That’s a further 6 quads of the current energy economy that we won’t need. Yes, we will have to build windmills, solar cells, batteries, nuclear plants, and electrical vehicles to substitute, but the energy used in their construction and operation is likely a smaller percentage of the 21st-century energy economy¹⁰. Solar panels produce the energy of their production in the first 1–2 years of their life, and last for 20–25 years¹¹. Wind turbines pay back the energy of their production in the first 6–9¹² months of a 20-plus year life.

Electrifying building and water heating saves us 5–7%.

Addressing how we heat water and air in our homes and offices is another huge opportunity for a new energy economy discount. Today, we typically use fossil fuels for heating our homes and commercial buildings, for providing heat for cooking, hot water, and clothes drying, and for providing heat to Industry. (In this case, the great majority is from natural gas and the remainder either fuel oil or biofuels.) But for “low-temperature heat,” meaning generally hotter than human skin but cooler than boiling water, we have an astounding and well-developed technology called heat pumps. Heat pumps are sometimes described as air conditioners that run backward, which is sort of true, but for this conversation all we need to know is that they are commercially available with coefficients of performance (COPs) of greater than 3, which means that for every unit of heat, they only need ⅓ of a unit of electricity.

Electrifying residential and commercial heat using heat pumps takes that sector from 7.5 quads of our national energy use to about 2.5 quads, another 5 quad saving. Low temperature heat used in industry uses another 3 quads. Electrifying that component via heat pumps would give us another 2 quad discount.

Without changing the size of our homes, or our cars, or fundamentally changing the fabric of our lives, these discounts mean that a fully electrified energy economy using non-carbon fuel sources would require less than half of the total amount of energy we use today.

LED lighting and Industrial Efficiency likely win us another 5%.

LED lighting, like electric cars, has matured greatly in quality, performance and availability. A wholesale commitment to LED lighting will save us another quad as lumen for lumen LEDs use only a fifth the energy of traditional lighting technologies.

More than 4 quads of our “energy flow” is primary fuels that wind up in “energy materials” like the coal in steel, energy embodied in fertilizers, wood used for construction, oil in asphalt, and oil and natural gas used in plastics. Some of these fuel uses don’t lead to emissions, and others will be replaced by innovative new industrial processes and biological sources of similar materials. There are yet more gains to be made in industrial efficiency.

In short, without changing the size of our homes, or our cars, or fundamentally changing the fabric of our lives, these discounts mean that a fully electrified energy economy using non-carbon fuel sources would require less than half of the total amount of energy we use today.

People will point out that you can’t electrify everything. That is absolutely true; for example a long-distance flight is very difficult to decarbonize, though it can be done with biofuels. America has an astounding biofuel resource that can be converted with known technology into natural gas, diesel and petroleum substitutes. For those pieces of the economy not amenable to electrification, and for reasons of seasonal storage to deal with winter heat demands, we probably need to double the amount of energy generated from biofuels from 4.5 Quads to about 9. This is a more than achievable number given the giant potential of US biofuels.

Some countries will struggle to produce enough zero carbon energy to satisfy their own needs; America however will have the opportunity to be a net energy exporter.

Where will the electricity we need in this 21st-century infrastructure come from? Today 38 quads of primary fuels delivers only 12 quads of electricity. In tomorrow’s fully electrified economy, we’ll need 33 quads of electricity delivered.

There are multiple paths to making all that electricity cleanly.

Is more than 30 quads of delivered electricity from renewables¹³ and nuclear¹⁴ a reasonable thing to anticipate? A conservative analysis of US roofs done by the National Renewable Energy Lab (NREL) estimates that rooftop solar could deliver about 6 quads of electricity. An ambitious conversion of rooftops would produce closer to 10–12 quads, with the advantage that this energy will be generated where it is used, eliminating the cost of transmission. Thinking outside the box for a second — covering 10% of the parking space in America with solar would produce more than 6 quads. America also has fabulous industrial-scale wind, biofuel, hydroelectric and solar resources, more than enough to satisfy another 20 quads. It is both feasible and reasonable to assume that the nuclear-powered electricity that currently produces 3 quads of our energy use from 100 plants could be tripled with next-generation nuclear to provide 10. Some countries will struggle to produce enough zero carbon energy to satisfy their own needs; America however will have the opportunity to be a net energy exporter.

With high certainty, we can say that in a decarbonized future, average US families will pay much less for all of their energy bills than they do today.

The details of the exact composition of the future grid is a detail that can be hammered out by policy, the market, and the people; the good news is that the target is more than reasonable with known technologies that in many markets are already by far the cheapest form of electricity and which are only getting cheaper. For example, rooftop solar in Australia produces electricity at around 7c/kWh for the consumer, approximately half the cost of the average retail utility electricity delivered in the US, which is 13.8c/kWh. With high certainty, we can say that in a decarbonized future, average US families will pay much less for all of their energy bills than they do today.

People will also point out that it will take energy to create the new 21st century decarbonized energy infrastructure. Solar panels, wind farms, electric cars, and heat pumps don’t grow on trees. During the transition period, the energy required to build this next generation of infrastructure will be significant. But the return on investment will be enormous. Reducing the total energy used in our economy by 50% means that these new investments will be profitable. They will create new industries, and will put millions of people to work.

We didn’t even mention some of the other advantages and wins that are possible. Short and mid-distance electric flight is not only possible but will make those flights safer, quieter, and use less energy. Insulating our homes and offices provides a smaller (traditional) efficiency win — or discount. Lighter-weight and more aerodynamic cars and a Cambrian explosion of small electric vehicles means that we might get further discounts still in transportation. Industry, particularly with new biotechnology providing alternative and better precursors for processing into a myriad of materials is another fabulous opportunity that the USA can be a leader in. Industrial automation and advanced manufacturing offers yet more opportunity.

This plan doesn’t address all sources of emissions. Important sources remain to be tamed, particularly the refrigerants used in air conditioning and refrigeration, and methane emissions from agriculture. There are technological solutions for these challenges too. I focus in this article just on the energy economy emissions for which I have done the most research.

There will be huge residual benefits in improved air quality, water quality, and quality of life. The health benefits accrued to us all will lead to large savings on healthcare for the nation.

America invented the auto loan in the 1920s and popularized the mortgage in the 1940s — these two inventions did more to shape the fabric of American society than any individual technology of the 20th century. Financing is like a time machine that allows you to have the future you want, today.

People will raise the specter of how we will pay for all of this. That is not the right way to think about it. Given the enormous cost savings from future efficiencies, the question should be “how do we finance this slam-dunk investment?” Reducing a major source of costs by 50% should be seen as the source of unassailable competitive advantage for our economy, an investment sure to repay itself many times over.

All of the energy technologies we have described above have high-capital costs but low operating costs. These are perfect opportunities for financing innovation. America invented the auto loan in the 1920s, and popularized the mortgage in the 1940s — these two inventions did more to shape the fabric of American society than any individual technology of the 20th century. Financing is like a time machine that allows you to have the future you want, today. If we could provide low-interest financing to this 21st century infrastructure in an analogous way to the low-interest financing we offered to consumers to buy cars and homes, and to utilities in the 20th century infrastructure build-out, this effort will be affordable to everyone and lock in lower and more predictable energy costs into the future. America can lead the world again in the banking and financing opportunity represented by the Green New Deal.

The Green New Deal is not naive in its ambitions or impossible in its scope. Quite the opposite, it represents the only viable pathway to American abundance and excellence since… the last New Deal.

Technological progress for the first time provides us with the opportunity to cleanly and cleverly provide all citizens with lower-cost access to all of the energy services they enjoy today and more. It is likely that this program has as its side effect a new American abundance. It is almost certainly going to improve the health and wellbeing of all Americans. It is also going to provide a huge new source of well-paid jobs. The installation for all of the electric heat pumps, solar roofs, electrical vehicles and other components of any massive electrification plan are largely physical, local jobs that cannot be off-shored. The appliance and automotive products that underpin this plan play to the strengths of American manufacturing. If America leads in the commercialization and manufacturing of these technologies it will provide for export opportunities that we haven’t seen since just after WWII when the rest of the world’s manufacturing infrastructure was decimated by war.

The Green New Deal is not naive in its ambitions or impossible in its scope. Quite the opposite, it represents the only viable pathway to American abundance and excellence since… the last New Deal


¹ https://www.eia.gov/totalenergy/data/monthly/pdf/flow/total_energy.pdf

² https://flowcharts.llnl.gov/commodities/energy

³ 1 quad = 10¹⁵ British Thermal Units (BTU) = 1.055 *10¹⁸ Joules = 172 Million Barrels of Oil Equivalent (MBOE) = 293,071,000,000 kWh = 8 billion gallons of gas.

⁴ https://www.innovationreform.org/wp-content/uploads/2018/02/EIRP-Deep-Decarb-Lit-Review-Jenkins-Thernstrom-March-2017.pdf

⁵ https://www.ethree.com/wp-content/uploads/2018/06/Deep_Decarbonization_in_a_High_Renewables_Future_CEC-500-2018-012-1.pdf

⁶ Throughout this paper, we sometimes use percentages instead of energy units, as for the purposes of public understanding this is the most useful way of contextualizing our energy challenge. Most large-scale energy planners in the US, and the agencies that collect the data use “Quads” which are Quadrillions of British Thermal Units, a not terribly intuitive unit. Fortunately, the US uses very close to 100 Quads, so for the moment anyway, 1 Quad is about 1 percent of US energy flow.

⁷ Curiously, and largely for historical reasons, the current reporting of solar plants, hydroelectric plants, wind and even geothermal, all assume generation losses equivalent to those from fossil fuel plants, so some of the 24% “lost” in generating our electricity is primary energy that in no real sense ever existed, or makes no sense to account for in analysis focused on emissions reduction instead of efficiency or energy independence. But whether we gain these back through a more accurate accounting of the benefits of other forms of electricity generation or through actual efficiencies, we can assume that the total amount of primary energy needed per unit of delivered electricity will be much lower in a decarbonized future.

⁸ For the truly detail oriented we currently calculate nuclear primary energy by the electricity produced via a heat rate. The heat rate is the efficiency of converting the heat of the nuclear reactions into electricity. It does not represent the `efficiency’ with which we burn the nuclear fuels — this varies enormously depending on whether uses light water reactors or breeder reactors. For the purposes of designing a zero carbon economy it probably isn’t useful to count the heat rate, rather only the delivered electricity from the nuclear power plants.

⁹ Hydrogen vehicles count too, as hydrogen is an energy storage technology just as batteries are. The hydrogen is generated by electric hydrolysis of water, and the hydrogen is turned back into electricity via a fuel cell that then runs the vehicle on electric motors. This may end up being key to long distance and heavy duty vehicles.

¹⁰ https://cleantechnica.com/2018/03/25/solar-power-energy-payback-time-now-super-short/

¹¹ http://www.iea-pvps.org/fileadmin/dam/public/report/technical/IEA-PVPS_Task_12_LCI_LCA.pdf

¹² https://www.researchgate.net/publication/240139630_Life-cycle_assessment_of_a_2MW_rated_power_wind_turbine_CML_method

¹³ For curious historical reasons, solar, wind, and hydroelectric production numbers have been assumed in the national energy accounting to have the same “inefficiency” as the average fossil fuel plant. This distorts the numbers and our perspective on the true advantages of renewables. We need to rethink this traditional measurement of “primary energy” in a decarbonized and electrified world. For this reason, we have switched here to talking about “delivered electricity” rather than using a flawed measurement of primary energy.

¹⁴ Today nuclear power’s “primary energy” is the amount of electricity delivered multiplied by the heat rate. This is really just a measurement of the efficiency of the steam turbine at the end of the plant, and ignores the 98% of energy left in the fissile material after reaction because we use for the most part light water reactors instead of breeder reactors.

Many thanks to Keith Pasko (who also produced the graphics), Jim McBride, Tim O’Reilly, Emily Leslie, Jonathan Koomey, Arwen Griffith, Patti Lord and others for their assistance in putting this together including useful commentary, edits, references, and feedback.

The Otherlab Blog

Otherlab is an independent research and development firm focusing on renewable and clean energy, robotics, automation, digital fabrication, adaptive textiles, advanced manufacturing, and computational design tools.

Saul Griffith

Written by

Founder / Principal Scientist at Otherlab, where he focuses his work on engineering solutions for energy production and energy efficiency. https://otherlab.com

The Otherlab Blog

Otherlab is an independent research and development firm focusing on renewable and clean energy, robotics, automation, digital fabrication, adaptive textiles, advanced manufacturing, and computational design tools.