Shorting Oil, Hedging Tesla

[neroden]

EV sales double every two years. Very roughly, oil demand displacement should go at the same rate (slightly higher, really). By BNEF's numbers it should reach 1.5 bpd per year in 8 years, or 2027.

Oil demand otherwise is increasing linearly, at 1.5 million bpd per year. So that makes 2026 the year for peak oil demand if BNEF has their numbers right.

 

[TheTalkingMule]

EIA is rapidly lowering demand growth forecasts, 2019 is now 1.45M and 2020 is 1.46. Avg will probably end up closer to 1.2 over the next 6 years, if that. Lord knows we're in for a slowdown.

 

[jhm]

Bloomberg about this topic:

Bloomberg - Are you a robot?

"Gasoline and diesel displacement by electric vehicles will grow by 96,000 barrels a day this year, BloombergNEF said in a report Tuesday. That brings the lost cumulative demand since 2011 at 352,000 barrels a day, about as much as total consumption by some countries, such as Peru or Portugal. By comparison, total global oil demand growth over the same period rose 12 million barrels a day to 100.6 million, according to the International Energy Agency."

So we still got quite a way to go...

I think BNEF may be underestimating displacement by a factor of 2.

 

[jhm]

EV sales double every two years. Very roughly, oil demand displacement should go at the same rate (slightly higher, really). By BNEF's numbers it should reach 1.5 bpd per year in 8 years, or 2027.

Oil demand otherwise is increasing linearly, at 1.5 million bpd per year. So that makes 2026 the year for peak oil demand if BNEF has their numbers right.

I'm a little more optimistic here than you are. IEA is projecting 1.2 mb/d annual growth average through the next 5 years. I also think BNEF is understating displacement by a factor of 2, but I'll ignore that here and play along with their numbers. I do anticipate 50% annual growth in displacement. Competition among Tesla, BYD and a few others will assure solid grow rates for a while as compliance cars and PHEVs fall to the sidelines in terms of growth.

So 350kb/d * 1.5^5 = 2658 kb/d cumulative in 2024 and 3987 kb/d in 2025. So this is 1329 kb/d incremental in 2025. This should suffice to quench positive demand growth. So I'm looking for a peak by 2024.

Even if BNEF is a bit off on their calibration, I really like that they are tracking this. Year after year, we can track this progress and update estimates of the peak. My expectation is that shifts in estimated peak will tend to shift earlier as time goes on. They will start to track other heavy EVs eventually which could show a big jump. Oil demand will also weaken as many factors erode demand. For example both China and India are pushing on rail and buses to avoid growth in auto. Ferries and heavy mining equipment are electrifying and on an on. So there is much more displacement and other efficiency gains than what BNEF (or anyone else) is tracking. In deed, to the extent that BNEF is underestimating EV displacement, there will be unaccounted for decline in oil demand. So my expect is that IEA and the like will be doing downward revisions to demand growth. So maybe a year from now, the five-year outlook may be more like 1.0mb/b rather than the 1.2mb/b today.

And finally, in my calculations above I went ahead and used BNEFs 350 number for 2019. That of course is a projection on their part. I believe it will be upwardly revised by this time next year. They are modeling just a 38% growth rate. Buses have nearly saturated, so this is suppressing their forecast. They are not looking at other heavy duty EVs. But auto EVs nevertheless are growing at a very fast clip this year. BNEF has put out this narrative that Tesla (i.e. auto EV) are not a threat to oil demand but buses are. I thing they are seriously missing the boat here. But their own projections incremental displacement in 2019 is 60 kb/p for buses vs 40 kb/p for autos, all while autos are growing 50%/y more or less. So it is not buses that threaten oil demand over the coming years, but rather autos. So I think BNEF was going after the provocative Tesla headline and was blind to where the action is going forward. It's not buses. It really is auto and other non-bus heavy EVs. In any case, I think they will catch up with this in a year or two. Let's see if 2019 cumulative displacement gets updated closer to 375 kb/d.

So I'm pretty bullish on peak oil demand by 2024, maybe a year or two sooner.

 

[jhm]

Survey confirms car purchases on hold as consumer interest turns electric

EV Osbourne Effect may be real, consumer survey shows.

Car buyers may actually be holding off on ICE purchases waiting for EVs. Global ICE sales are down, but direct consumer research is helpful for clarify why this is the case. Interest has definitely shifted away from gasoline vehicles and toward PHEVs and BEVs.

In terms of impact on oil demand, I suspect that the decline in ICEs is actually more important than EV fuel displacement. The peak size of the ICE fleet makes a big difference in total CO2 emissions from vehicles. The smaller the ICE peak is, the more quickly EVs can replace them. So an EV Osborne Effect could be the most important impact on climate change that EVs can make at this time.

 

[neroden]

I'm a little more optimistic here than you are.

I did say if BNEF has their numbers right. I suspect BNEF has displacement low and oil demand growth high. I mean, my model still says peak demand in 2023. Their numbers imply 2027.

Even if BNEF is a bit off on their calibration, I really like that they are tracking this.

Me too

 

[jhm]

I did say if BNEF has their numbers right. I suspect BNEF has displacement low and oil demand growth high. I mean, my model still says peak demand in 2023. Their numbers imply 2027.


Me too

Yes, it is interesting to point out the implications of their numbers. It is also curious that BNEF does not spell this out. Early on in the demand peak story, BNEF was doing really good work to show just how quickly all this could transpire. Since then, they've been back peddling on anything that tweaks the nose of the status quo, so called "conservative" estimates are the order of the day. For example, BNEF is still targeting 2024 as the year when battery costs will drop below $100/kWh, the magic threshold that final make EVs a thing. I found out how they get to that. They base it on and experience curve with an 18% rate, but what they do is assume that the growth rate of EVs will slow down, which via the experience curve slows down the rate at which the price declines. So they are creating a kind of circular logic wherein assuming that EV growth will slow for lack of lower prices leads to a delay in lower costs which leads to slower growth of EVs. Why is this so convoluted? If you simply assume that current rates don't slow dramatically for the next couple of years, you get to the magic number by 2021. So their conservatism is really an excuse to impose an abrupt slowdown of current trends. As a data analyst my basic motto is to let the data speak. I can impose my own biases disguised as reasonable assumptions on data all day long, but at some point I've got to step back as ask what is the data telling us. From every angle we've looked at this question about the rise of EVs and impact on oil demand, it has been clear that trends indicate a rapid transition. It's actually hard to find data that suggest that the transition is slowing down. So yeah, just look at the "conservative" analyses BNEF is putting out, connect the dots, and oil peaks by 2027. But BNEF refuses at this point to clarify just how quickly oil could peak.

My frustration is with BNEF and so much more media. The peak denialism is so thick. At risk, the oil industry need to invest $10T over the next 20 year to perpetuate their BAU. Investors are starting to become suspicious of the long-term oil story and demand returns and avoid potential stranded assets. The value of this industry is totally dependent on this flow of capital. But they are facing a crisis of confidence. In this context, BNEF backs off on its own estimates and analyses that that suggest rational grounds for this crisis of confidence. It's not healthy. When the peak does come more damage will have been done simply by the fog of denial obscuring better choices.

 

[neroden]

Based on the combination of leaks and available pricing data and targeted profit margins, I think Tesla's below $100/kWh at cell level already. Pack level is below $150/kWh, and I'm more certain of that. The fact that it's all gone hardcore trade secret makes it harder to tell, though.

 

[neroden]

Yes, it is interesting to point out the implications of their numbers. It is also curious that BNEF does not spell this out.

Yeah, that was my point. You're right that they seem to be deliberately sandbagging their articles. Why?

So yeah, just look at the "conservative" analyses BNEF is putting out, connect the dots, and oil peaks by 2027. But BNEF refuses at this point to clarify just how quickly oil could peak.

Why? With BNEF, this is a real question. It's not like GTM/WoodMac; they have no oil clients.

 

[jhm]

Big Pivot In Energy Is Gaining Momentum | OilPrice.com

 

[jhm]

Yeah, that was my point. You're right that they seem to be deliberately sandbagging their articles. Why?



Why? With BNEF, this is a real question. It's not like GTM/WoodMac; they have no oil clients.

Actually, BNEF does consulting on renewables and energy transition stuff. So they very well be toning their reporting and analyses so as to preserve "credibility" with potential clients. Not sure I like the conflict of interest between reporting and consulting.

 

[TheTalkingMule]

Never thought about the possibility of oil majors pulling an Aramco and purposely trying to bury the Permian independents. Is this what's happening?

From oilprice...... A Paradigm Shift In The Permian | OilPrice.com

It certainly smells like the small frackers are teetering. Permian production is on a straight vertical growth pattern and the najoma can afford to eat losses for a bit. Is there logic in this idea they would keep drilling knowing full well they're gonna drive WTI to $40 where they lose money?

 

[RobStark]

One concern that I have had for a number of years is that the price of ICE vehicles would decline just as EVs become competitive and scale up. If this were the case it would make it harder for EVs to capture market share profitably.

However, a 1.5% profit margin for Nissan suggests that they are in no position to cut prices on ICE just to defend market share. I suspect other ICE makers are in the same situation too. Production of unprofitable models will be cut.

In 2009 Nissan had 6.5% US Market Share.

In 2018 Nissan had 9.35% US Market Share.

They have gained share by cutting prices and margins.

The Detroit 3 having been doing the reverse. Cutting unprofitable small cars and sedans. Cutting unprofitable sales to Rental Car companies.

 

[jhm]

From California to Oslo: foreign subsidies fuel Norway's e-car boom, for now

This is interesting. Norway seems to have an insatiable demand for EVs, both new and used. Perhaps lawmakers wishing to incentivize EVs should look at incentives that accrue with use post-sale. For example, a carbon tax on fuel coupled with subsidized charging would give preference to operating an EV over and ICE. A simple tax credit on the purchace of a new EV does little to incentivize retention of that vehicle as a used vehicle.

I think as we get further along in the EV transition managing the used car inventory will become a bigger issue. I actually think that exporting EVs is a good thing particularly where it makes EVs more affordable to less wealthy markets and helps to preserve resale values in the more affluent markets.

One wonders, is Norway exporting used ICE vehicles?

 

[mblakele]

From California to Oslo: foreign subsidies fuel Norway's e-car boom, for now

This is interesting. Norway seems to have an insatiable demand for EVs, both new and used. Perhaps lawmakers wishing to incentivize EVs should look at incentives that accrue with use post-sale. For example, a carbon tax on fuel coupled with subsidized charging would give preference to operating an EV over and ICE. A simple tax credit on the purchace of a new EV does little to incentivize retention of that vehicle as a used vehicle.

I think as we get further along in the EV transition managing the used car inventory will become a bigger issue. I actually think that exporting EVs is a good thing particularly where it makes EVs more affordable to less wealthy markets and helps to preserve resale values in the more affluent markets.

One wonders, is Norway exporting used ICE vehicles?

This happens within the USA too: Californians buy used EVs from out of state that are eligible for new HOV stickers, and export used EVs to other states. Since there's a general increase in the number of EVs in California, there's probably an outflow of used ICE to other states too. All this activity is basically non-productive, unintended consequences of the existing incentives. A sanitized carbon tax would be much more efficient.

 

[jhm]

To be sure, one good thing that exporting used EVs does is that it help to support resale values. If California was not able to export these used EVs, the prices for a used EV would drop. This would in turn create a disincentive to buy new EVs. In terms of fighting climate change, it matters little where an EV is driven (although Norway does have very low carbon electricity). What matters most is that demand for new EVs is strong and supports global scale up. So exports of used EVs supports the global scale up.

Think of this example. Germany did a really great thing by heavily subsidizing solar before it was cost competitive. Sure, there are sunnier places on earth than Germany that would have generated more solar power per panel. But what Germany did was push solar down the experience curve rapidly at a time that it needed the boost. California and Norway are doing a similar thing for EVs.

 

[jhm]

How Big Oil Could Become Big Electricity | OilPrice.com

I could say something snarky, but I yield to ennui.

 

[jhm]

Mexicans embrace bikes, scooters as fuel shortage drags on | Reuters

Here's a bit on how electric scooters have helped to curb the impact of gasoline shortages in Mexico. The obvious point illustrated here is the willingness of those who use autos (private and shared) to substitute scooters when the price (including wait time finding available gas) goes too high. Basic economic thinking would have anticipated this, but it is good to see it in action.

The point is there are consumers that are at the margin between using a car and using other micromobility to meet some of their transportation needs. The explosion of electric scooters could be the tipping point for many. A combination of electric scooters, public transit and occasional Uber trip may well suffice to avoid an increment purchase of a car.

Could electric scooters and bikes be materially contributing to sagging auto sells? It would be nice to size this up. The real problem for oil demand growth is not so much growing EV sales as it is declining ICE sales. So we should be looking at any alternatives that might materially impact ICE sales.

 

[jhm]

Uber sees a 38% yearly increase in gross bookings as it gears up to face competition

An analysis that came out last month from Santhosh Rao, a policy researcher at Uber, pointed out that Uber trips in San Francisco have fallen by 10% since February, which also marked the entry of Uber into the scooter-sharing market. The number of taxis hailed fell even harder during peak hours, where Uber trips went down by nearly 15%.

So here we see that electric scooters can make a material cut into Uber ride-hailing service. We need to keep this in mind as we contemplate the prospects for robotaxis. The competition is not merely between private auto ownership and ride hailing, but also with other forms of micromobility. I do see micromobility and ride-hailing as complementary, however, The combination of the two make private auto ownership less attractive. The two work together for a lower cost, higher availability alternative. Also in terms of eroding oil demand, the combination of the two leads to even lower per mile energy consumption than ride-hailing alone.


The Electric Scooter Fallacy: Just Because They’re Electric Doesn’t Mean They’re Green
Here is one attempt to analyze the carbon emissions of electric scooters. I think the author is working way too hard to reach his biased conclusions, however. The energy use for the scooters themselves is 6 to 14 Wh/mile! That is just 2% to 5% of the energy use of an EV. The whole beef with emissions, rather, revolves around the logistics of charging. Driving many miles in a van to recharge does produce some serious emissions, but the this is just brown washing the issue. Regularly I see collectors moving 3 to 5 other scooters on one scooter. This is to charge and place the scooters. If charging location are near enough for collection on other scooters, then these logistical miles cannot be much more than about as many rented miles put on the scooter. So worse case your 14 Wh/mile rent incurs another 14Wh/mile. So we still get to some trivial energy use like 30 Wh per rented mile. The point is that as the market for rented scooters matures some very low cost and low energy logistics can emerge. And of course as EVs advance, EVs running mostly on renewable energy will be used as logistical support for electric scooters. So the concept is entirely compatible with net zero carbon economy as are larger EVs. (I really despise the brown washing of EVs and other clean tech on the grounds that they currently participate in dirty fossil fuel driven economy that browns up the supply chain of f'n everything. The question should be how a technology is compatible with a low carbon economy. That my pet peeve.)

 

[jhm]

Goldman Sachs: U.S. Drillers Aren’t Hedging Enough | OilPrice.com

What could go wrong with that? OPEC could break up!

​