Are the Inflation Reduction Act's Climate Goals Plausible?

A 40 percent cut in U.S. greenhouse gas emissions by 2030 is possibly achievable.

The Inflation Reduction Act (IRA), hammered out through negotiations between Senate Majority Leader Chuck Schumer (D–N.Y.) and Sen. Joe Manchin (D–W.Va.) and unveiled last week, has been hailed as the "most important climate action in U.S. history." Why? Because it "reduces carbon emissions by roughly 40 percent by 2030" below the level emitted by the U.S. in 2005, according to a press release from Senate Democrats.

The consulting firm Rhodium Group, after cranking the Act's provisions through its econometric model, concurred: "Our preliminary estimate is that the IRA can cut US net greenhouse gas emissions down to 31% to 44% below 2005 levels in 2030…compared to 24% to 35% under current policy." The policy shop Energy Innovation's analysis agreed: "We find that the IRA is the most significant federal climate and clean energy legislation in U.S. history, and its provisions could cut greenhouse gas (GHG) emissions 37-41 percent below 2005 levels."

U.S. greenhouse gas emissions fell to six gigatons in 2020 from their 2005 peak of 7.4 gigatons, a 19 percent drop. Reducing greenhouse gas emissions to 40 percent below their 2005 levels implies cutting another 1.6 gigatons of emissions in the next seven years or so. Does this seem plausible? Neither the Rhodium Group nor Energy Innovation's black box analyses break out how this would be achieved.

So let's make some very rough calculations to see that if that goal is at all plausible. First, consider the four categories of GHG tracked by the Environmental Protection Agency: fluorinated gases (largely coolants), methane, nitrous oxides, and carbon dioxide.

Fluorinated gases: In May, the U.S. Senate approved the Kigali Amendment to the Montreal Protocol that phases out fluorinated gases by 2036. Under the American Innovation and Manufacturing Act, the U.S. commits to cutting fluorinated gas production by 70 percent by 2030. This should roughly cut its carbon dioxide equivalent emissions from 189 million tons to 57 million tons—that is, by 132 million tons.

Methane: The IRA includes a carrot consisting of $850 million in incentives to monitor and mitigate methane leaks and a stick imposing a fee of $1,500 per ton of methane leaking from oil and natural gas facilities. At the Glasgow climate change conference last year, the Biden administration promoted the Global Methane Pledge, in which 121 countries agreed to cut their methane emissions 30 percent below their 2020 levels by 2030. In 2020, U.S. methane emissions amounted to the carbon dioxide equivalent of 650 million tons. Cutting that by 30 percent would reduce those emissions by nearly 200 million tons.

Nitrous oxides: Agricultural production is responsible for 74 percent of U.S. nitrous oxide emissions, mostly from the application of organic and synthetic fertilizers and raising livestock. It will take lots of different relatively small-scale interventions to reduce nitrous oxide emissions. Nevertheless, let's make the heroic assumption that the $20 billion allocated by the IRA to the agricultural sector can help meet the same target set for methane emissions. That is a 30 percent cut of agricultural sector nitrous oxide emissions by 2030. Such a cut would amount to the equivalent of about 95 million tons of carbon dioxide.

So adding up prospective cuts of fluorinated gases, methane, and nitrous oxides totals a GHG reduction equivalent to about 430 million tons of carbon dioxide. That would mean that cuts in carbon dioxide emissions must be around 1.2 billion tons by 2030. It is worth noting that U.S. carbon dioxide emissions fell by 1.4 billion tons between 2005 and 2020. In other words, to achieve the goal of cutting overall greenhouse gas emissions by 40 percent by 2030 suggests that a roughly equivalent reduction in carbon dioxide emissions would have to take place in about half the time. In 2020, most U.S. carbon dioxide emissions arose from electric power generation (1.4 billion tons) and transportation (1.6 billion tons).

A 2021 analysis of power generation trends by the U.S. Energy Information Administration calculated that the carbon dioxide emissions in that sector fell 32 percent from 2.5 billion tons to 1.7 billion tons between 2005 and 2019. Over that period, the percent of power generated by coal fell from 50 percent to 23 percent, whereas the portion produced by burning natural gas rose from 19 percent to 38 percent. When burned, coal produces about twice as much carbon dioxide as does natural gas. The agency attributes 65 percent of the decline in carbon dioxide emissions to switching from coal to natural gas generation. About 30 percent of the decline resulted from increases in renewable power generation.

Modeled pathways for reducing carbon dioxide emissions vary, but let's run through an exercise using specific greenhouse gas emissions numbers to see what might be possible. First, posit that by 2030, coal-powered generation is phased out while keeping natural gas generation. If coal-powered generation is entirely replaced by renewable power generation, that would cut carbon dioxide emissions by 770 million tons. So additional cuts amounting to 430 million tons would be needed to get to the goal of reducing greenhouse gas emissions by 40 percent below their levels in 2005.

A March report from the American Public Power Association (APPA) notes that U.S. electric utilities have a total nameplate generation capacity of 1.2 million megawatts. Nameplate capacity is the maximum electrical generating output of a power facility. Coal accounts for 18.5 percent (232,000 megawatts) of that capacity, wind 10.7 percent (134,000 megawatts), and solar 5.3 percent (67,000 megawatts). However, since wind and solar energy is intermittent, it is not possible to simply switch out one-to-one coal generation for renewables capacity.

Given current electricity market conditions, the capacity factor of renewables is about half that of coal. So replacing coal would require building out around 460,000 megawatts of solar and wind generation capacity by 2030. APPA projects that about 30,000 megawatts of new solar and wind capacity will be deployed this year. Replacing all coal-fired generation would require more than doubling (65,000 megawatts) that rate of deployment for solar and wind power beginning next year. That is about the rate China achieved last year.

Again, even if coal power generation is completely phased out, 430 million more tons of carbon dioxide emissions would still have to be cut in order to reach the goal of a 40 percent reduction below their 2005 levels.

The largest source of greenhouse gas emissions in the U.S. is the 290 million cars, trucks, and buses on American roads, emitting about 1.6 billion tons of carbon dioxide in 2020. Notionally, to cut transportation emissions by 25 percent (400 million tons), that would mean that a quarter of the fleet in 2030 would have to be powered by electricity. Assuming a fleet of 300 million, some 75 million electric vehicles would then be on U.S. roads. One estimate projects roughly 50 million electric vehicles traveling U.S. highways by 2030. Perhaps the IRA subsidy of $7,500 per electric vehicle would be enough to entice Americans to buy that many electric vehicles between now and 2030. In fact, Bloomberg is reporting that the U.S. has just passed the inflection point where electric vehicles amout to 5 percent of new car sales.

Powering millions of electric vehicles would likely require the generation of more electricity, which would impact greenhouse gas emissions in that sector. On the other hand, managed charging and discharging to the power grid while vehicles are sitting idle in driveways and parking lots might be one way to address the problem of renewable power intermittency.

Of course, these very rough calculations do not take into account the costs and the inevitable regulatory roadblocks (we'll leave that to another time). But the modeled claims that it is possible to reduce U.S. greenhouse gas emissions 40 percent below their 2005 levels seem plausible, if just barely.