Every winter I teach Sustainable Design at The Creative School and the School of Interior Design of X University, these days mostly about reducing carbon emissions. Right now I am discussing decarbonization. We have covered many of these themes on TreeHugger, but it could be a useful roundup.

It is fashionable these days to rally round the cry “Electrify Everything!” or as British engineer Toby Cambray recently verbed, “Heatpumpify Everything!” The promoters of this idea are led by entrepreneur and inventor Saul Griffith, who wrote a report that starts with the bang.

It is an interesting and seductive idea, and it is based on sound reasoning: We have a carbon problem, not an energy problem. So if everything is electric and all the electricity is low- or zero-carbon, problem solved! Use as much as you like—same–sized homes; same–sized cars. Just electric. Buy two if you like.

The problem, before we even start looking at how we decarbonize, is to figure out how much carbon we are talking about, to wrap our brains around the size of the problem.

We know that if we are going to keep the global heating below 1.5 degrees Celsius (2.7 degrees Fahrenheit) there is a ceiling on the amount of carbon dioxide (CO2) we can add to the atmosphere—420 gigatonnes at the time this chart was made. It’s a lot lower now.

We have what’s called an “emissions gap” where we are emitting roughly 55 gigatonnes of carbon dioxide equivalent (CO2e) every year and by 2030 we have to reduce that to about 22 gigatonnes per year, a reduction of 32 gigatonnes per year. The numbers aren’t as dire if you are aiming for a 2 degree Celsius increase but I am not willing to go there yet in these discussions.

So where are all the CO2 emissions coming from? This EPA graph of U.S. emissions divides it up into sectors, but it is really not so simple. It seems that we spend an awful lot of time talking about buildings when, according to the EPA here, they only total 13% of emissions.

The lovely Lawrence Livermore Lab Sankey chart of CO2 emissions shows much the same thing, with the bulk of the carbon emissions coming from transportation, electricity, and industry.

When you look at where energy is coming from and where it is going, (Full-size image here) the picture changes dramatically. Sixty percent or 21 quads of electrical energy are coming from coal and gas, and that has to switch to low or zero-carbon sources very quickly. (A quad is a quadrillion British thermal units or BTUs.) About 75% of the electricity is going into our residential and commercial buildings, 9.34 quads, and 60% of that (5.6 quads) is dirty. This is ignoring all the rejected energy going up the chimney; this is real BTUs that are used.

About 45% of the natural gas (8.08 quads of energy) is going straight into our buildings. Gas furnaces are probably averaging 85% efficiency, so that is delivering 6.86 quads of useful heat. If we get that out of heat pumps with an average Coefficient of Performance year-round of 3, that’s 2.286 quads of energy. So that means that to electrify our buildings, we have to generate 7.88 quads of new clean electricity, which is about half again the 15.3 quads of solar, nuclear, hydro, and wind power that we have now.

This is why I keep banging on about efficiency, but also about sufficiency, about not building more than we need; that’s a lot of quads to find in a hurry, almost all of which is going into heating and cooling.

And we haven’t even started talking about the industry. According to Our World in Data, steel is responsible for about a third of industrial emissions, and half of that is going into buildings and infrastructure. That’s roughly another 5 quads of clean energy needed.

Cars are another story. They use up a vast amount of energy, more than any other item on the Sankey chart, but they are incredibly inefficient, using only 5.09 quads of energy to move the car with the rest going out the exhaust or as wasted heat. Electric cars are far more efficient: According to Natural Resources Canada, “The efficiency of energy conversion from on-board storage to turning the wheels is nearly five times greater for electricity than gasoline, at approximately 76% and 16%, respectively.”

So getting 5.09 quads of useful driving would need only 6.69 quads of electricity. Most people do not have to fill their electric car batteries at peak times, so it is likely that operating electric cars won’t actually add more than half of that to the peak load scenario, so now we only have to find 11.2 quads of clean reliable green energy to power our buildings and cars. But that is also why getting rid of gasoline-powered cars is the single most effective way to reduce carbon emissions.

At this point, it almost seems like I am making the argument for the Electrify Everything! gang; 11.2 quads don’t sound so hard to find, just 10 times all the solar or 4 times all the wind power we have now or increasing nuclear by 50%. Easy!

But we have to build all those cars out of steel and aluminum, with upfront carbon emissions between 10 and 40 metric tons per vehicle. There are 276 million cars registered in the U.S. Replacing them will emit a very big burp of carbon. That’s why we shouldn’t just focus on electric cars, but on getting by with fewer cars, and figuring out how to get around without them.

The point of all this is that we can’t just electrify everything and believe in the “same-sized everything, just electric” train of thought. Operating residential and commercial buildings, building them, and building and operating the vehicles and infrastructure to get between them, are probably close to three-quarters of our carbon emission pie. There just isn’t enough low-carbon juice to power it all. We have to radically reduce demand as well as electrify everything.

That’s why we have to change the way we live, we have to change the way we work, and we have to change the way we get around.

More to come.