Shorting Oil, Hedging Tesla

[jhm]

All of the actual adoption curves by country or region follow the "S" curve postulated by Tony Seba (and others) to arrive at the estimate of 100 m vehicles by 2030. I don't think it is overoptimistic to think this trend won't continue.
ICE will be in serious trouble in just a few short years.

Yeah, that sounds like Seba. I do hope he right. But realistically I don't think the last 10% of the market matters that much. ICE and oil are both toast if EVs are near 90% my 2030.

There is a kind of symmetry in the logistic curve, the least complicated S curve. EVs were at 14% in 2022. If it takes four years to get to 50% EV, 2026, then 4 years later, 2030 non-EVs hit 14% (86% EV). That's good enough for me. But if EVs hit 50% in 3 years, 2025, then 86% EV can come by 2028, and 95% by 2030. Even better!

My point here is simply that the time it takes to get to 50% is critical. That's also the time of maximal change in market share. I have over the years though it would come around 2027. But the solid performance of Chinese EV makers makes me a more optimistic that 2026 is possible.

Also I don't think Chinese EV makers will slow up all that much as EVs hit 50% of the Chinese market. Rather, they will turn to export market to sustain high annual growth rates, even as growth in domestic demand slows. This is one reason why I don't worry too much about country specific S curves. For me it's the global market that matters.

So let's see when EV will hit 50% of the market. How soon do we see that coming?

 

[gavine]

Is it safe to say that Norway is a microcosm to what the world will be by 2030? I realize Norway accelerated their adoption with heavy incentives, but I feel like they are a proof of concept that EVs can work for the masses.

 

[mspohr]

Is it safe to say that Norway is a microcosm to what the world will be by 2030? I realize Norway accelerated their adoption with heavy incentives, but I feel like they are a proof of concept that EVs can work for the masses.

Some people say that Norway is a special case but there is a growing list of countries following the same path.

 

[jhm]

Is it safe to say that Norway is a microcosm to what the world will be by 2030? I realize Norway accelerated their adoption with heavy incentives, but I feel like they are a proof of concept that EVs can work for the masses.

Incentives in Norway put them ahead of the curve in terms of relative cost to ICE. However, they are not really ahead on product diversity and product quality. As the rest of the world catches up in EV adoption, consumers will have even better choices than are currently available.

For example, the US market has a strong demand for trucks, but there are too few EV truck offerings at the present moment to address this segment demand in volume. This is why we're all eager to see Cybertruck ramp up.

Had Norway comparable demand for trucks, it probably would not have as much EV penetration as it currently does.

I think, when it comes to global EV penetration, we need to see continued product development in every segment. The top sellers in every segment need to be BEVs.

Oh, yeah, and prices need to keep coming down too.

 

[jhm]

Here's a thread worth reading.

https://twitter.com/x/status/1722219871306695149

 

[jhm]

This is quite interesting. Li-ion really looks to be the most important storage tech for daily balancing, and hydrogen for seasonal balancing.

https://twitter.com/x/status/1722544993179504965

 

[Oil4AsphaultOnly]

This is quite interesting. Li-ion really looks to be the most important storage tech for daily balancing, and hydrogen for seasonal balancing.

https://twitter.com/x/status/1722544993179504965

Does cloudless AND windless days span more than 24 hrs? Doesn't the chart essentially mean that hydrogen is never a competitive option and will always lose out to compressed air?

 

[jhm]

Does cloudless AND windless days span more than 24 hrs? Doesn't the chart essentially mean that hydrogen is never a competitive option and will always lose out to compressed air?

Yeah, depending on how much overcapacity of wind and solar, there can be many days in which the 24-hour generation of wind and solar is insufficient to fully recharge batteries. If you've got a lot of battery capacity, you can run a deficit for many days in a row. But at some point you need some kind of backup generation to keep your batteries from being fully discharged.
One modeler in Australia has been modeling a system with 5 hours of storage (24GW × 5 h = 120GWh) and enough wind and solar to cover 105% of annual demand, he finds that this covers all but about 1.2% of power delivered. This Other generation needs a peak capacity of about 7GW, maybe 5% CF. Hydroelctric storage is also in the mix. He thinks hydrogen could provide this critical backup need.

https://twitter.com/x/status/1722077906040762612

So I do think there will be places where hydrogen is needed, but it is a little more complex than just a storage function. Electrolyzers provide dispatchable load on on hand, while a hydrogen powered turbine provides dispatchable power on the other hand. Both sides contribute to stabilizing the grid. If Australia had enough wind and solar to cover 200% of non-electrolyzer demand for power, that could be enough to avoid virtually all need for back up generation. Every day there is a surplus of wind and solar generation, and dispatchable electrolyzers soak up this surplus. So that's at one extreme. At another extreme electrolyzers have very limited capacity and just barely enough to provide backup fuel for hydrogen generators. This extreme is hydrogen as grid storage only. I am more skeptical about grid storage only being a cost effective solution. It also does not address other decarbonization needs like green steel or green ammonia. So to my thinking, if electrolyzers are substantially helping to decarb outside of the power grid, they will largely do seasonal balancing of the grid by virtue of subsidizing oversupply of RE for the grid.

I'm not sure how this storage researcher would disentangle these two sides of the hydrogen question. This is also why researchers should be exploring decarbing the whole energy system, and not just power grids in isolation.

 

[jhm]

https://twitter.com/x/status/1722077910633505061

These charts are pretty important. Notice that there is 20% overgeneration, not 5% as I stated above.
Note also that peak other generation (backup) is about 9GW with 2.7% CF. This is very peaky, but the peaks happen mostly during the winter (down under).
You can see the obvious problem of solar energy creating seasonal balancing issues, not enough sunlight in winter. Wind helps a bit, but the week to week variability of wind is very high.

Basically, it looks like north facing solar to optimize winter production would be a big help for seasonal balancing. I don't think David's data nor modeling are able to represent that sort of alternative supply.

I think we can expect that as solar reaches higher levels of penetration in a grid more attention will be given to winter, sunset and sunrise generation. Those harder to supply hours will be what makes or breaks the financial performance of incremental solar.

Oh, yeah, wind developer need to work on diversification to minimize week to week variation in production.

If this is true in Australia, then we should see David's setup do an even better job minimizing backup generation than it does on current data where solar and wind are not yet optimized for high penetration. This is very optimistic that 5 hours of battery storage may be quite enough in this part of the world.

 

[ggr]

Yeah, depending on how much overcapacity of wind and solar, there can be many days in which the 24-hour generation of wind and solar is insufficient to fully recharge batteries. If you've got a lot of battery capacity, you can run a deficit for many days in a row. But at some point you need some kind of backup generation to keep your batteries from being fully discharged.
One modeler in Australia has been modeling a system with 5 hours of storage (24GW × 5 h = 120GWh) and enough wind and solar to cover 105% of annual demand, he finds that this covers all but about 1.2% of power delivered. This Other generation needs a peak capacity of about 7GW, maybe 5% CF. Hydroelctric storage is also in the mix. He thinks hydrogen could provide this critical backup need.

https://twitter.com/x/status/1722077906040762612

So I do think there will be places where hydrogen is needed, but it is a little more complex than just a storage function. Electrolyzers provide dispatchable load on on hand, while a hydrogen powered turbine provides dispatchable power on the other hand. Both sides contribute to stabilizing the grid. If Australia had enough wind and solar to cover 200% of non-electrolyzer demand for power, that could be enough to avoid virtually all need for back up generation. Every day there is a surplus of wind and solar generation, and dispatchable electrolyzers soak up this surplus. So that's at one extreme. At another extreme electrolyzers have very limited capacity and just barely enough to provide backup fuel for hydrogen generators. This extreme is hydrogen as grid storage only. I am more skeptical about grid storage only being a cost effective solution. It also does not address other decarbonization needs like green steel or green ammonia. So to my thinking, if electrolyzers are substantially helping to decarb outside of the power grid, they will largely do seasonal balancing of the grid by virtue of subsidizing oversupply of RE for the grid.

I'm not sure how this storage researcher would disentangle these two sides of the hydrogen question. This is also why researchers should be exploring decarbing the whole energy system, and not just power grids in isolation.

Why hydrogen, though? It's hard to store long term. Why not make methane instead?

 

[jhm]

Looking more closely at this, minimum weekly production of solar is about 30% of weekly demand. The week where backup power was most needed hit 11% of weekly demand. Basically, the grid needs about 50% more solar in winter weeks.

Rooftop and utility solar are calibrated to provide 45% of annual demand. Boost this to 68%, and it should come close to eliminating the need for backup.

Also that weekly variation in wind is just brutal. Sometimes one week is 40% of total demand lower than the week before. That is a huge amount to smooth out via storage. Hydroelctric storage only compensates a little. Mostly the high wind production one week is just surplus.

By contrast solar provides fairly stable amounts from one week to the next. Boosting winter generation would make it even more reliable through the whole year.

 

[adiggs]

Why hydrogen, though? It's hard to store long term. Why not make methane instead?

Yeah - "hydrogen" seems like an umbrella term to me, either for technology and materials that don't exist, or for some other chemical formulation than H2. Ammonia, Methane, I don't think H2O works all that well But something mixed in with the hydrogen to stabilize it at least a little bit.


As I understand it, even hydrogen used in fixed industrial processes is generated and consumed on the spot. With the possible exception of rockets I don't believe that anybody is trucking around large quantities of liquid hydrogen, or storing vast quantities of H2 for later use. If for no other reason a truck load of liquid hydrogen getting in an accident on the highway could make for a really sweet bomb.

Heck even the hydrogen refueling stations, at least some of them, used grid electricity to make hydrogen on site, rather than relying on a tanker truck to swing by now and again and refill the underground tanks.


EDIT to add: Hydrogen storage - Wikipedia
Lots of good info here. Not as much in this article about just how difficult it is to keep liquid hydrogen in a container, or the downsides of uncontrolled release. But lots and lots of options being researched regarding other stuff that will take up hydrogen, and also release it later in an easy and controlled fashion.

EDIT again to add: Underground hydrogen storage - Wikipedia
Much more hopeful to push hydrogen gas into underground storage than I thought. It is something we've been doing for decades, so its well past lab technology. It wouldn't be portable - the facility that makes the gas would be pumping, and extracting, the hydrogen gas directly. But for grid storage application, portability isn't needed.

 

[jhm]

Why hydrogen, though? It's hard to store long term. Why not make methane instead?

Hydrogen is just the starting point. Methane is good for using existing gas infrastructure. Ammonia can be shipped as a liquid. These options are reasons why hydrogen is more than just a way to store energy.

 

[RobStark]

Falling demand for gasoline is reducing profitability of gasoline.

 

[Doggydogworld]

Crack spreads have been insane the past two years. Old, grimy refineries became literal cash registers. Just crazy. The current 15-20 cents/gal spread is pretty typical for this time of year, historically speaking. Let's hope things have finally normalized for good.

Here's US gasoline consumption the past 30 years. A steady climb into the 2007 housing bubble then a decade-long lull until ZIRP and cheap fracking finally pushed us back up to nearly 10m bpd summers. Then COVID and WFH wrecked the curve and 2022's $5 gas hit demand. Will prices fall enough to get us back to 10m bpd next summer? Or maybe summer 2025? Your guess is as good as mine.

 

[jhm]

Chart of the Day

I like the idea of tripling clean energy investment by end of decade, but I'm not sure what this means for total energy investment.

Consider that recently clean energy investment has been about $1.80 per $1 invested in fossils. Suppose we multiply this ratio by 2.6x. Thus $4.7 clean per $1 fossil. This is a nice ratio, but to get the absolute level of clean investment up 2.6x then total energy investment must be about 2.03x the current level (= (1.8×2.6 + 1)/(1.8+1)) if fossil investment remains at the same level!

Obviously, fossil spending must decline. In the most extreme it drops to virtually 0. Even here total energy investment climbes 1.67x (= 1.8×2.6/(1.8+1)).

Thus, total energy investment needs to increase by 70% to 100% by 2030 to keep 1.5C climate goal on track. While I am absolutely bullish about increasing the ratio of clean to fossil investment, what worries me is, how does the global economy marshal the motivation to nearly double total energy investment? Does this not imply that most of the fossil investments happening this decade will yeild negative returns if we stay the course? That is, when their is overinvestment in energy, this leads to oversupply, a glut which implies horrible ROI for most investors.

 

[jhm]

It may be helpful to express the above in annual growth rates. Increasing clean energy investments 2.6x over the next 7 years is 14.6% per year. If fossil investment drops to 0 over these 7 years, total energy investment is up 1.67x or 7.6% per year. If fossil investment remains constant, total investment is up 2.03x or 10.6% per year.

Increasing energy investment 10.6% per year seems awfully rich and destined to produce a glut. A glut of gas, oil and coal would have to lead to serious reduction in fossil investment. Worse, clean energy investment would also fail to grow 14.6% per year. The planet would be flooded with cheap fuels stalling clean investment.

Even 7.6% annual growth in total energy investment seems like a serious risk of oversupply. But if we assume that and 14.6% growth in clean, we can back into a residual fossil investment curve.

2023 1.00
2024 0.95
2025 0.88
2026 0.78
2027 0.65
2028 0.48
2029 0.27
2030 0.00

This represents fossil investment as a ratio to 2023 levels. You can see that the bottom really falls apart in the last couple of years of the decade. It's forced to go to zero investment levels because clean investments are oversupplying all energy markets.

Under this scenario I think there is generally too much total energy supply building up. So fuel price collapse much earlier on. Suppose for example that total energy investment can grow no faster than 6% per year without a fuel price collapse. This gives a shorter curve.

2023 1.00
2024 0.91
2025 0.78
2026 0.63
2027 0.43
2028 0.19
2029 0.00

With this curve, fossil investment ends more quickly, and with less cumulative investment. This is advantageous to fossil investors because it does a better job of maintaining fuel prices. It generates better ROI on fossil investments, while putting less carbon into the atmosphere.

Is this aggressive enough? I don't know. I think oil investors need cut investment levels fast enough to keep Brent north of $80, while gas and coal investors need to keep gas prices in NA north of $4/mmBtu. And clean tech investors need these high energy prices too to sustain 15% growth in investment levels.

Basically, I see energy prices in the hands of fossil investors. I am confident that clean investment can rise to meet the gap faster than fossil investment can drop off. You see silly arguments like oil producers can't just stop producing oil this year. True enough, they would only do that if they were in the midst of a massive glut. But to reduce oil investment by 10% to 20% in a year? Sure, we could see tighting in oil supply and oil could treble north of $100/b. In such a scenario, EVs double each year while oil reserves are drawn down. Oil investors get massive ROI. We get to equilibrium in a couple of years. No problem.

The global economy does not collapse because of underinvestment in oil. OPEC needs to stop being such a drama queen. All fossil investors need to ramp investment levels down to 0 over the next 5 years. Go ahead and starve the economy for fuels and feast on big, fat returns. Clean energy can and will rise to the challenge.

 

[jhm]

Chart of the Day

https://twitter.com/x/status/1723936276900905046

I like the framing here. In India, solar with enough battery capacity to store 50% of generated energy is competitive with solar. But this is just in terms of average cost. The stored solar power can be discharged during the hours that would be most profitable for coal. Coal is slow to ramp up and down, while batteries can respond immediately to price signals. This flexibility would make solar plus 50% battery storage potentially more valuable, more profitable than coal for the same average cost.

I suspect that 50% storage may be close to economically optimal for the bulk of solar installations. Basically, it nearly optimizes the CF of inverter and interconnection, around 50% with lots of flexibility. 100% would have even more flexibility, but not discharging while the sun shines would underutilize the storage capacity. Thus 100% is suboptimal with too much capacity. And of course, 0% storage underutilizes inverter and interconnection capacity. Thus, the optimal amount of storage capacity is between 0% and 100%. The midpoint of 50% is a good place to look for a maximum. But if we consider the ratio of battery capacity to inverter/interconnection capacity, as this ratio declines, the optimal mix will shift toward more storage.

Coal wants to run some 16 or more hours a day (2/3 CF). When there is sufficient solar, coal operates on the 16 hours that has the least sun. Solar with 50% storage generates solar direct to grid about 8 hours. The other hours of storage competes with coal on the most profitable 8 of the 16 went coal is burning. These best 8 hours must be above the average cost of coal, otherwise coal becomes uneconomical, too much coal capacity on the grid. And this is exactly the point. Solar plus 50% is competitive enough to displace coal capacity.

 

[jhm]

IEA Raises Oil Demand Outlook For 2023 And 2024 | OilPrice.com

Global oil demand continues to exceed expectations, with the International Energy Agency raising its demand growth forecasts for 2023 and 2024.

oilprice.com

We frequently react to downward adjustments to oil demand forecasts. Today we have an increase to consider.

IEA is bumping up there 2023 demand increase from 2.3 mbpd to 2.4 mbpd to hit 102 total for 2023. Okay a rounding error, not bad from a forecasting point of view.

Also boosting the 2024 increase from 0.90 mbpd to 0.93 mbpd, with 2024 demand to total 102.9 mbpd.

Basically this is just a small fine-tuning. What I find significant here is that of the 2.4 increase this year, China accounts for 1.8 of that. Implying, as oil demand growth in China halts, global demand growth becomes a rounding error.

Next, consider 2.4 growth in one year followed by just 0.9 the next. That is huge pullback in growth. It also means that China is not expected to grow much, certainly not 1.8mbdp as in 2023.

China is at the very center of the demand peak drama. IEA modelers now see that growth in China's demand is no longer reliable. The structural element is massive growth in EVs specifically in China. And there is also a random component here that depends on how strong the economy is in China. Any slow up in economic growth will disappoint oil demand forecasts. This economic component also means that near China's oil demand peak, we may see decline one year followed by growth the next, even as EVs gobble up market share.

 

[jhm]

https://twitter.com/x/status/1724106858397352258

Also, China looks to enter structural decline in carbon emissions in 2024. This is mostly around solar and wind holding coal at bay. But any slowing of oil demand in China also helps the move into structural decline.