Plugging the Coming Jevons' Carbon Leak

Jevons Paradox applies across energy types which means that for cheap solar energy to reduce carbon consumption, we need a floor price on oil.
The above sentence pretty much summarizes this entire article but is almost incomprehensible, so now I'm going to spend a stack of words unpacking it.

Jevons Paradox Explained
Jevons Paradox states that when you have a resource that is useful for all sorts of stuff, the more efficiently you can use it, the more of it people will use. The classic example and the one that is relevant for our conversation is energy. If cars get 1 mile per gallon, people only drive when they absolutely must and very sparingly when they do. If cars get 100 miles a gallon, people don't worry about how the gas cost part of their commute. If cars get 1000 miles a gallon, people look for new creative ways to attach little car engines to anything they can and don't even bother putting in an off switch.

The more efficient gas is, the more it gets used for everything. You can see similar effects happen with computation power; as the cost of a FLOP drops, people buy more of them, so many more that they spend more on computing than they did before. You can also see it happen with money; when a startup is uncapitalized, they rarely try to solve problems with money but the more cash they get, the more they solve problems by throwing money at them.

Jevons Paradox Applies Across Energy Types
I never put it together until it was pointed out that Jevons Paradox can hop between similar resource types. Though there is a little bit of friction, if the price of one kind of energy goes down, the cost of other types of energy will go down to stay competitive, which means that there will be more consumption. For instance, when we as a culture moved to gasoline oil, it didn't reduce the use of coal. It just dropped the price. Peak coal usage was actually in 2009, and it only started going down as a result of regulatory action.

J.M Greer pointed this out here:

What makes Jevons' Paradox so deadly in the present situation is that adding new sources of supply to the energy mix has the same effect as making demand more efficient. That's why it's inaccurate to claim, as so many badly written histories do, that oil replaced coal. More coal gets burnt each year now than was burnt each year at the peak of the coal era; petroleum, by taking some of the demand that would otherwise drive up the price of coal, kept coal cheap and made it economical to use coal even more freely than before.

In exactly the same way, adding wind and solar to the energy mix doesn't replace fossil fuels. As shown by the energy use chart above, it takes some of the demand that would otherwise drive up the price of fossil fuels, and thus keeps the price of fossil fuels lower than that would otherwise be. As a result, more fossil fuels get burnt.

Solar is on track to be much cheaper than gas.
A few things have shaken me out of my general climate change pessimism lately, and one is that Solar is getting really cheap. So cheap that it's already cheaper than gas if you count infrastructure costs. So cheap that it's going to start driving the cost of gas down soon. The drop in the price of solar is almost entirely good news. Very cheap energy carbon-neutral energy will lift the standard of living for everyone, it's going to allow for some cool tricks like carbon capture and desalinization, and it will open up the possibility of displacing oil consumption.

But we need to make sure that we don't fall into the trap of Jevons Paradox with it. The current floor price on gas is lower than the cost of pumping it from the cheapest wells and refining it. Why lower? A few reasons, because huge reserves are held that cost money to store and would probably need to be shed. Because of sunk cost into infrastructure. Because of quirks in Byzantine financial and legal structures that assumed gas would pump forever.

That's a problem for world carbon output.
As a species, we want to get carbon use down, and Jevons Paradox is a place where we will lose a significant advantage. I'm pretty excited about carbon reduction in ways that don't make any large group of people suffer or ask them to make massive sacrifices in growth or progress. I'm pretty sure that we will end up having to ask people to make enormous sacrifices. Large groups of people are going to end up suffering before climate issues are resolved either way, but it would be good if we could get some of the less painful stuff done first as a first step.

It's tough to pass taxes and laws that eat into existing business.
In the US and most of the world, it's hard to change laws in ways that challenge an existing business model. The political will generally can't beat the lobbyists, and even when it can, it's usually pretty challenging. In this case, it's quite difficult to pass a tax that increases the cost of gas because several large industries will oppose that tax. In short, Exxon doesn't want you to make their product more expensive and will actively oppose any attempt to do so. But there is a trick:

It's much easier to pass a law that closes off a version of the future that hasn't happened yet.
Suppose you happen to have a different forecast from the entrenched businesses. In that case, you can probably push laws through that will change things in the world that you are forecasting without a lot of opposition. In this case, I'm anticipating that the cost of gasoline will drop p to a low price per gallon.

It takes about half a barrel of oil to generate a MWh of energy. Right now, a barrel of oil costs ~$60, making a MWh cost around $30 on existing infrastructure. So what happens to the price of oil when the cost of a MWh from solar is around $10? I forecast that the cost per barrel goes down to approximately match the price of solar, and we see something like $20/ barrel.

After all, OPEC+ isn't going to stop pumping just because the price is dropping; if anything, we have seen that they pump more in response because their member countries are structurally dependent on the income from oil sales. This scenario isn't some far-off sci-fi future. It's not crazy to guess that solar will cost $10/MWh by 2040. But as far as I know, few are looking at it that way.

Here's my proposed law:
A tax that effectively sets the minimum price of a barrel of oil to $35. If the oil price is at or above $35 it does nothing, there is no tax. If the price drops to $34, the tax is $1/ barrel. If the price drops to $20 the tax is $15/ barrel. The dollar denominations there need to be indexed to inflation.
Oil costs $60/barrel right now and hasn't had a stable price below $35 since the early '70s. Many people are acutely aware that we are running out of oil anyway and expect the price to increase.

Does that law sound minor and irrelevant, and stupidly small? Good, it's supposed to! It's intended not to raise any resistance. It's supposed not to cause any new pain but still provide a much-needed spur to switch to solar when the price of solar is easily viable.

I would love to see stronger laws come in first. I would love to see cap and trade, a carbon tax, or just about anything coherent and more aggressive. Of course, this is a desperate backstop law that only matters if the US government can't get its ducks in a row in the next 20 years, but honestly, that seems like a possible future to me, so I think that it's worth considering.