82 Comments

Good post!

One corollary is that we should be working to make electricity as cheaply and widely available as possible, right? I have a friend who’s replacing his home HVAC system and is debating between sticking with gas or going with a heat pump. In California, our electricity is so incredibly expensive, and the rates increase with usage, that gas is competitive. It shouldn’t be that way – the goal should be to have electricity be like tap water – i.e., not something you really think hard about using too much of. (And it doesn’t have to be that way – Arizona’s electricity is about 30% per KWh compared to California.)

Another point: the limiting inefficiency of electricity is transmission over distance. If you take what you’re saying a step further, a lot of our use of electric power in the future will be to power data centers to perform computational tasks. Digital information, however, does not suffer from transmission loss over distance. I’d think it would make tremendous sense, and there’s a huge opportunity, (and I’m sure people are working on this…) to co-locate energy production with data centers in remote locations – say, the remote desert, where solar generation is more efficient, but where the loss from electrical transmission over long distances erodes the gains from efficiency.

The point is that the continuing digital revolution enables you to take a lot of the “work” to the optimal place for electrical generation, rather than the usual problem we’ve faced of how to get the electricity to the place where the work happens.

Expand full comment

Based on your post, I checked out some AZ utility rates and -- as a PG&E customer -- it almost made me cry. Not only are the fully-bundled volumetric rates much lower, but the off-peak time-of-use rates are a fraction of PG&E's. The "super-off peak" rate (11pm-5am) for APS is less than 4 cents/kWh, which means you can charge a 4 mile/kWh EV for less than a penny per mile, about 1/8th the best PG&E rate.

Part of it must be lots of low-cost nuclear from Palo Verde, which was commissioned in 1986, as well as a good amount of cheap hydro. Together these comprise about 35%(?) of AZ supply, with cheap gas making up another 30+%.

Expand full comment

Yeah, it’s bonkers. We bought our in-laws’ place in the desert near Tucson when they moved into a retirement community… I had to look at the first power bill a half-dozen times before it sank in. It was so much less expensive, I thought I’d missed something.

I don’t know what’s driving the price difference. But it’s incredible, and at some level the cost and unreliability of California’s electricity has to be a drag on California’s economy.

Expand full comment

I live in Tucson for the last 18 years. TEP prices have doubled and especially after the foreign country Fortis energy bought them out

Expand full comment

CA _grid_ electricity is expensive, but that makes it incredibly valuable to get enough solar and batteries that you don't have to pay for grid electricity. I basically just carry a credit balance on my electrical with PG&E. :-P

I am mildly annoyed that they've messed with net metering enough that exporting power to the grid has almost no value, but whatever. I'm somewhat seriously considering just going off grid. I have a friend who's done that. I do like having the grid as a fallback option in case all else fails, and leaking somewhere between $10 and $30 of electricity onto the grid each month, that they'll never pay me for, is kind of a small price to pay for having that last resort, which is unlikely to consume the amount of negative balance I've accumulated if I have some kind of problem for a few days at some point. (If I did want to go fully off-grid, while still finishing off electrifying everything on my property, and reliably never having a period where we'd exhaust the storage, I'd probably need to roughly double my current PV and storage. But I do have the roof space for that.)

I'm also somewhat intrigued by the new generation of home microturbines. Considering trying to integrate a TESUP 9kW vertical turbine, as an additional resource, to supplement my PV and PowerWalls.

Expand full comment

A friend of mine has a huge amount of solar and battery. To the point where he doesn't really care about insulation in his house anymore, which is an interesting side effect of energy abundance I hadn't thought about before. He also has the heating on all night so he never has told deal with cold floors in the morning.

Expand full comment

Love the energy abundance implications!

Expand full comment

Can you explain more what you're talking about about with the data centers ? I get electricity, and transmission, but I'm pretty shaky on digital technology and computation.

Expand full comment

Sure. Let me try: Electrical transmission over long distances is difficult because electricity is difficult to transmit without significant losses. (This is part of what makes superconducting materials interesting, i.e., that you can send electricity across them with very little loss of energy). Data transmission over long distances is robust (i.e., you don’t lose data) and uses relatively little electricity.

If you want to power a data center in a population center with solar, and you want to build a giant solar farm in a remote desert with a lot of sun (e.g., Death Valley, etc.) where the conditions for solar power generation are optimal, you have a problem because you can generate a lot of energy, but you still have to ship the electricity to a population center, which will be incur a lot of inefficiencies from transmission. However, because data moves more efficiently over distance than energy, you can build a data center in the desert next to the solar farm, “ship” your compute loads over a distance to the data center. You could apply this approach to any location where there’s a clean energy source but where few people live, to optimize energy usage. For example, geothermal energy, etc.

If an analogy helps: It’s like a mill driven by a water wheel… you could build a mill in town and divert water via an aqueduct to turn the mill, or you could build a mill next to a river. Let’s say that the farmers have to harvest the grain in the field and transport it somewhere, regardless. That’s work that is going to be done irrespective of where you put the mill. So all else being equal, it will be more efficient to build the mill next to the power source, and do the work there. Same thing with the data center example: the fact that data can be transmitted efficiently means that you can take the work to where the power is generated, rather than bringing the power to where the work is done.

Expand full comment

ah i see thank you for your response, that makes sense.

Expand full comment

Yeah Noah! Finally, you have returned to optimism. This is a very interesting framing of this technological moment. The problem with electrifying the high income countries is the concentration of resources in the hands of the combustion caucus. That makes me pessimistic about the rate and extent of electrification which can be achieved in the US. I'm doing my part since my last two vehicles have been electric. However, I don't hold any fossil fuel stocks.

Expand full comment

Noah,

Great article. Probably the best since your 2008 market crash stuff. The concept of leap frogging existing technologies is fascinating and examples are already out there. Large majority of phone communication in Africa skipped landlines and went right to cell phones. Cheaper, faster and better in their case.

It will be interesting to see if one of our really big energy companies takes the leap. Unfortunately, unless battery technology makes a tremendous leap, which is possible, power distribution is the big issue. We have truly MAJOR sources of both wind and solar, BUT not where it is needed. Based on my somewhat limited 2nd hand experiences, the transmission of electricity is at least as much a political issue as it is a technological issue. We have nowhere near the transmission capacity we need now (brown outs during the east coast heat wave), much less enough capacity for the electrical future we both see coming. It takes investment capital, which we do have, AND political capital, which we do not seem to have.

If China needs a transmission line, they build it right through whatever neighborhood gets in the way. We take years of meeting and then court cases.

Anyway, keep up the 21st century blogging.

One addendum. Fusion will always be 30 years in the future. Go with the sun. It worked it out billions of years ago.

Expand full comment

Thanks! I started blogging in 2011 though, so you might be thinking of someone else. :-)

Expand full comment

Not if you started blogging when you were at UM grad school in Econ, and were blogging about econ and the stock market. Mainly I remember Noahpinion. Don't think there are two of you, but time does fly. I remember the crash in 2007-8 and trying to figure how I missed understanding the madness. I guess I didn't catch your act until 2011, and was still trying to figure out how it happened.

Expand full comment

The same argument is why I’m skeptical that fusion power will be economically viable.

If a fusion power plant is just another heat source for a steam turbine, I fail to see how it’s going to be cost-competitive with the solar and battery technology of a couple of decades from now.

Expand full comment

Check out aneutronic fusion! If that ever works, it'll be amazing.

Expand full comment

Is this the Helion pitch?

Expand full comment

No idea if they'll be the ones to do it (I suspect not). But it'll happen eventually.

Expand full comment

I think we’ll eventually realize we need fusion to fail due to the waste heat problem.

Expand full comment

The main reason why nuclear fission failed to live up to its promise was the rise of the LNT-regulatory-industrial complex, and wouldn't that also sabotage fusion power even if the immense technical challenges of fusion could be overcome?

Expand full comment

Fusion power isn't dangerous at all so highly doubtful

Expand full comment

In theory, fusion could become competitive because it's just so much denser than any other energy source. Fusion of hydrogen to helium converts 0.6% of an object's mass into kinetic and thermal energy, which may not sound like a lot but the equivalent for fossil fuels is 0.00000006%, ten million times less.

The only problem - you know how the main inefficiency of fossil fuels is trying to harness useful work from the chaotic release of thermal energy. Now try to imagine doing that with an actual nuclear warhead...

Expand full comment

Noah - What do you think the US should be doing to become a major producer of solar cells and batteries? IOW, how can we take away the existing dominance that China has right now? How can we make American citizens - especially the ones serving in Congress - see that this is really an issue of national security and act accordingly? Are we totally f***ed if Trump gets back in power?

Expand full comment

I get that electromagnetism vs thermodynamics gets more clicks than electricity vs combustion, but it's not like electricity is free from thermodynamics and combustion is free from electromagnetism. See data centers where huge amounts of cooling are needed—and actually, combustion is a chemical process meaning it's essentially explainable through electromagnetism and quantum mechanics (I guess this is just saying electromagnetism is a lot more than electric currents). </pedantry>

Expand full comment

Check out the alt-text on the pic

Expand full comment

Calling it a metaphor is fair, and that's why I had the pedantry tag. But to continue my pedantry, the reason (classical) electromagnetism can be violated is due to quantum effects, which are stronger when you're looking at high energies/small distances. This is distinct from how thermodynamic laws can fail when you have small numbers of particles because the the theory is statistical in nature.

Expand full comment

My partner’s research is specifically on the thermodynamic limits and possibilities of converting between light and electricity, so I was particularly bemused by this proposed as a contrast! Solar panels are heat engines too, except your working fluid is light rather than steam (and your hot side is the temperature of the sun, rather than the puny temperature of a blast furnace).

Expand full comment

Maxwell made major contributions to thermodynamics too.

Expand full comment

"My worry is that the U.S. and other developed countries won’t follow this approach, and will cling to increasingly outmoded combustion-based approaches. I’m not sure they’ll realize how broad and sweeping the shift is, or how unlikely it is to be reversed."

The "realization" would happen a lot faster with a tax on net emissions of CO2 and removal of all the other high-cost measures to reduce net CO2 emissions and of the regulator obstacles obstacles to development of net zero CO2 emissions technologies.

Expand full comment

Don't hydrocarbons still supply the majority of electricity and aren't China and India building many coal plants alongside solar as sources of electricity?

if so, then are you comparing combustion to electricity or non HC sources to HC sources? I was confused.

Expand full comment

You get energy from hydrocarbons by burning them. So they always involve combustion.

Expand full comment

No, not always, though in the vast majority of applications. There are fuel cells based on hydrocarbons or carbohydrates that do precisely controlled non-combustion generation of electricity. Yes, they are based on redox reactions, but combustion is only one class of redox reactions.

Expand full comment

Fuel cells provide approximately 0% of electricity production (and other forms of industrial energy) and that does not look set to change in the near future.

Expand full comment

Quite right. Like I said, vast majority. But they have long been considered as an alternative to combustion: a technology that hasn't panned out but still could in the future.

Expand full comment

I had the same skepticism of the RMI chart for China and/or globally. I wonder if they are describing end use energy modality (fossil or electricity) for energy services at the consumer (output) level, not on a primary energy * efficiency basis on the input side. I enjoy RMI, but think they present data in a fast and loose way oftentimes.

Expand full comment

https://ourworldindata.org/grapher/co2-by-source?country=CHN~USA

For "the first electrostate", China sure is doing a lot of combustion of coal.

I'm not yet sold on this theory. The technologies Noah is describing require difficult to obtain elements often located in hostile places and highly complex production processes. They may be efficient (and I think there's some rose-colored glasses going on there too) but they are not resilient. And the last 5 years should have taught us the importance of supply chain resilience, particularly of key technologies like generating power.

Expand full comment

Not that difficult to obtain elements actually and much lower quantities of them, fossil supply chains are far more prone to disruption as we've seen many times throughout history.

Also China is burning so much coal (as opposed to nat. gas and oil) for the exact same reason they are investing so heavily into renewables and EVs. They are the largest importer of oil globally and hugely vulnerable to supply chain disruptions and blockades.

Expand full comment

I agree about the reareason, but I think Noah's claim that China is "the first electrostate" is at a minimum optimistic. They burn more of the dirtiest fuel on the planet than any other country, and are building more capacity to do the same faster than anyone else.

Oil and coal and natural gas are all fairly straightforward mining and refining processes.. The supply chains are only global for cost. They could all be insourced easily.

Expand full comment

Due to geographic distribution of oil and gas deposits, this is not true at all for most countries. If it were true, 70s oil crisis wouldn't be a crisis, OPEC wouldn't be a thing etc.

Expand full comment

I was talking about for the United States.

Expand full comment

This is a really fun post and I have basically no qualms with the subject matter, but the language I would argue is slight off. For example, I wouldn't say that combustion is understanding the physics of "random motion," or that combustion wastes energy through the "randomness of heat."

In reality, heat isn't random at all; we can predict where it will go and why with a high degree of accuracy. The dilemma is that it is devilishly hard to route heat with any degree of efficiency, and it is nearly impossible to do when you get to a small enough scale.

To a lesser degree I also found myself wincing at the implication that electricity is a heatless resource that always goes precisely where you want it to, but that is because I am in an industry (medical devices) where our lack of ability to do so is a daily struggle.

All told if I read this on another forum I wouldn't have blinked, but Noahpinion has enough precision of language that I couldn't help myself from commenting :)

Expand full comment

Heat is random at the micro level, which electric current is not! We predict the behavior of heat at the ensemble level...

Expand full comment

Well wait, is this an article about fascinating truths of physics at the micro level, or what is true on a measurable level that define which of electricity or combustion is winning out?

Because on a functional level, again, heat is not random in any meaningful way that is driving why electricity is winning. We can (and do) predict how much heat will be generated and when and where it will go with a *significant* amount of accuracy. In fact, in modern high-precision devices, the problem is *far* more that we failed to predict where the electricity would go rather than failed to predict where the heat would go, even in devices that operate based on generating heat.

Again, I'm not saying that your actual points were wrong, or even that your specific wording was imprecise. I'm "merely" saying that this reads like someone that isn't used to writing in this space.

Expand full comment

If we ever develop a truly limitless source of energy waste heat will become an existential issue.

Expand full comment

Although I agree that Asia probably has an advantage in the electrification of everything space, I think the electrification of Bangladesh and Vietnam is overstated. People in Bangladesh and Vietnam probably just drive fewer kilometres per capita than Westerners (still dominated by ICE). Plus their households invest more home and office cooling (already electric) than heating systems (gas based for now).

Expand full comment

Heating in East and Southeast Asia where it exists is also overwhelmingly heat pump.

When I was visiting northeastern Thailand, the room I was staying in had a heat button on the air conditioner remote.

In Tokyo, people refer to heating as "aircon" just like cooling as heat pumps for heating have been the norm since the 2000s if not earlier. The non-loanword word for an air conditioner is a "reiboudanbou" literally meaning coolerheater.

Expand full comment

Haven't you warned us in the past against charts with arrows pointing up?

Expand full comment

https://www.technologyreview.com/2024/03/07/1089585/hydropower-trouble-droughts/

China is chasing the loss of hydropower due to droughts with permitting two new coal-fired power plants each week.

Climate change-induced extended droughts and record heat waves is accelerating the world’s largest dirty coal deposits.

Expand full comment

The abundance agenda is really promising, but I have a couple of concerns

* Abundance of resources doesn't guarantee that everyones needs are met. The world has long produced more than enough food for everyone, and could produce more, yet a billion people are still going hungry

* The version being pushed by Steve Teles and Rob Saldin includes plugging nuclear power, the ultimate in heat-based technology (not strictly combustion, but its appeal is that it does everything combustion can, on a bigger scale).

Expand full comment

Efficiency of electric motors didn't improve that much during the 20th century, they have been pretty good for a long time, even brushless motors provided just about 15% improvement. Battery energy density however quintupled since the beginning of 20th century, so that's probably the biggest difference maker in electrification of transport.

Expand full comment