North America

Devil’s Advocate: An unconventional, long-shot case for Elon Musk

I would not invest in Tesla. I think Elon Musk’s style is a little bit annoying, and I think many of his supporters are very annoying. More importantly I am not sold on the claim that Tesla will be able to compete against other auto or tech firms, even assuming that electric vehicles really do become widespread soon.

Looking at Musk’s business moves individually they appear, at best, to be high risk, high reward.

For example:

— Tesla’s approach to autonomous driving is not to use LIDAR, because LIDAR is expensive. This is unique: the other auto and tech firms are all betting on LIDAR. And because the economic viability of electric vehicles probably depends on autonomous driving (the vehicles need to be able to drive themselves to and from charging stations, as otherwise charging batteries may be too inconvenient when compared to conventional or hybrid vehicles), if this LIDAR-free strategy fails, it might put Tesla in a very tough position.

— Large electric trucks do not seem to make obvious economic sense: the batteries are too big, bulky, and expensive. It is difficult to see why these would be able to compete, in the short run, against conventional trucks, and in the long run against robots making it much easier to transfer cargo between electric railways and “first-mile/last-mile” conventional trucks or smaller electric trucks.

— Solar City. Even assuming that solar can compete with other power industries, and even assuming that using batteries to store power can compete with other forms of energy storage, it is difficult to see why a diffuse system of rooftop solar panels would be able to compete with solar farms, where installation and maintenance costs per panel are lower and where there is less shade.

–Boring company. Even assuming that Musk does succeed in reducing urban tunnelling costs, such tunnels would still be hugely expensive, so it is not clear why you would use them to move cars or people on `sleds`, when it would be much more efficient from a capacity point of view to simply use an existing technology within the tunnels: namely, trains.


The Unconventional, Long-Shot Case: Tesla Parking Lots 

Readers of this blog will know I have a weird obsession with parking lots, because parking lots are the most ubiquitous type of American real estate and because they may be impacted more than other types by technologies like e-commerce and autonomous parking. Let’s imagine what Elon Musk might be able to do with a typical supersized suburban parking lot:

— No LIDAR, no liability, no problem: while autonomous vehicles in general might need LIDAR and might face liability issues, in a controlled, pedestrian-free environment — for example, in a designated autonomous zone of a parking lot — an autonomous car could function without LIDAR. This would have two benefits: one, it would act in effect as a valet service, making it easy to park; two, the parking lot could have an autonomous charging station for electric cars, so that your car could be charged while you are in the mall

—  Sledding. The car-carrying ‘sleds’ imagined for use inside the Boring Company’s tunnels may not make economic sense within those tunnels, but they could make sense as  sleds that could carry conventional, non-autonomous cars (there are hundreds of millions of these cars in America today, and they aren’t going to disappear overnight) to and from parking spots.

— The Boring Company. If the Boring Company ends up reducing the cost of conventional subway trains, the value of autonomous valet parking lots could increase, as people will drive their car to a parking lot at the nearest subway station, then get on the subway train while the car goes to park itself. (They may also be able to get in another car at their destination station’s parking lot, thus overcoming the ‘first-mile, last-mile challenge’ that plagues suburban transit in America today). Short-distance tunnels created by the Boring Company could also be used to link together parking lots that are close together: lots of suburban parking lots are giant ‘archipelagos’ separated by highways, for example.

— Electric Trucks. Electric trucks may not be economical in general, but could be economical in a specific situation: driving short ‘first-mile/last-mile’ distances, in daytime or overnight (electric vehicles are quiet, so better for nighttime use) between, for example, a commercial/industrial parking lot and a rail or conventional truck logistics station. So, for example, a company like Walmart could use electric trucks to bring in cargo quietly at night when its parking lot is empty, and also charge their batteries in the lot.

— Solar City. Rooftop solar panels may not be economically competitive in general, but on large flat roofs with little shade — notably, on large commercial/industrial roofs, next to large parking lots — they may be more economical. It may even become economical to put a solar roof above the large parking lots, to generate power while also helping to keep the parked cars shaded.

Okay, I admit, this is all unfounded, unclear, and far-fetched. Ultimately, it is based on the assumption that if wholly autonomous cars do not become widespread in the near future, then the most efficient, clean, and convenient methods of transportation and commerce may instead involve a combination of electric cars, conventional transit, and autonomous parking. Elon Musk’s unique mix of assets may be uniquely suited for this outcome.

Advertisements
Standard
North America

Night Moves: The Future of Charging Electric Cars in Ontario

The Ontario government recently announced a plan to subsidize electric cars by up to $14,000 per vehicle and pay for them to be charged at night, among other things. Night-time charging is a key factor in electric vehicle ownership, as in most cases it takes several hours to charge an electric car.

This begs the question: what will the price of overnight electricity in Ontario be in the years ahead?

Today overnight electricity is cheap because most nuclear power plants in Ontario and coal plants in nearby states like Michigan cannot easily be turned off at night, in contrast to gas plants or hydropower facilities which can more easily ramp up and down their output to match real-time electricity demand. An estimated 60 percent of Ontario’s power is generated from nuclear, compared to around 15 percent in Canada as a whole and 20 percent in the US. Around 50 percent of the power in Michigan, Pennsylvania, Ohio, Indiana, and Illinois (states that are close to Ontario) comes from coal, compared to 33 percent in the US as a whole and 10-15 percent in Canada. Ontario and the Midwest are also among the leaders in wind turbines, which do not turn off at night either, and Ontario, Illinois, and Pennsylvania are by far the top three North American producers of nuclear power.

Going forward, however, there are compelling reasons to think that this overnight surplus of electricity will no longer exist.

The first reason is fracking. In the past few years the US has seen an enormous boom in shale gas production, which has been leading much of the country to begin switching off their coal plants and replacing them with cheap natural gas. The stock prices of US coal companies have already dropped by over 90 percent since 2014, and by over 97 percent since 2011. As more gas and less coal is used to generate electricity, the price of overnight electricity is likely to spike relative to the price of daytime electricity, since gas plants tend to be far easier to shut off at night than coal plants.

This is relevant to Ontario because the biggest gas booms in the US since 2010 have been in nearby states like Pennsylvania, and because Ontario already has the extensive natural gas infrastructure required to import and distribute American gas (especially via Michigan, which has the largest gas storage capacity in the US). Indeed fracking has made gas so cheap in the region that even Ontario might look to it again as a source of energy production, instead of building new nuclear plants or wind farms.

The second reason overnight electricity prices are likely to rise is robots. Machines that combine mobility with computation are highly energy-intensive, but, unlike humans, they do not need to sleep at night or relax in the evening. Take, for example, Amazon’s robotic warehouses: they have caused the company’s night-time electricity use to rise substantially since they were introduced, given that before they came along Amazon’s warehouses were either inactive overnight or else employed human workers who ran on food (and overtime pay) instead of electricity. If and when this robotic economy finally goes mainstream, then, such demand for overnight power could be replicated at large. We should expect late-night electricity use to skyrocket: robots are no longer science-fiction.

The final reason is environmentalism. In order to keep greenhouse gas emissions down (which is, after all, the main point of subsidizing EVs), many voters are pushing for more solar panels and wind turbines to be built. Solar and wind are complementary to one another, not only because the sun often shines brightly at different times as the wind blows strongly, but also because wind farms and solar farms usually inspire non-overlapping types of NIMBY-driven political backlash. Ontario already gets 5-6 percent of its electricity from wind compared to less than one percent from solar, so it might be that going forward its solar power growth will outstrip its wind power growth. Of course, solar power will not help to bring down overnight electricity prices. Even the wind, however, tends to blow less strongly overnight than during the day – a fact that runs contrary to conventional wisdom, since the wind can usually be heard more clearly at night.

As solar, wind, and gas replace dirtier coal in the regional electricity network, there will also be environmentalist-led pressure to stop heating homes with fossil fuels and instead adopt electric-powered heaters like those used in Quebec and the Pacific Northwest. This too would be likely to cause overnight electricity prices to rise. Quebec, for example, uses electric-powered heating and so has its electricity demand peak during frigid winter nights, whereas Ontario primarily uses gas-powered heating and therefore has its electricity demand peak during hot summer days. Should Ontario or nearby US states switch over to electric heating in order to reduce carbon and methane emissions from natural gas, the region’s overnight electricity usage will rise.

The need to help support solar and wind power could lead as well to the building of more pumped-hydro facilities, which pump water uphill so that it can flow back downstream through a turbine when other power sources are in low supply, such as when solar panels are blocked by clouds or the wind is not blowing. There has been talk lately of building more pumped hydro in Ontario, in places like Niagara and Marmora, as pumped hydro is the most efficient form of electricity storage. Given that Ontario’s daytime power is not cheap (at least, not by Canadian or American standards), this water would be pumped at night. It is an energy-intensive process, however, requiring 20 percent or so more energy to pump uphill than is generated from releasing it back downhill. Thus it would lead overnight prices to rise.

In closing, any electric-vehicle policy approach that assumes that Ontario’s overnight electricity costs will remain cheap is probably a shortsighted one. Ontario’s overnight electricity costs are likely to rise substantially as a result of natural gas replacing coal, robots working slavishly every night, and the move towards cleaner sources of energy like wind power and, especially, solar power.

Without being certain of future electricity prices,  the EV subsidy plan is like a leap, or Leaf, in the dark.

Standard