Battery technology may emerge as a trillion-dollar threat to credit markets
bloomberg.com
bloomberg.com
"I used to think if there was reincarnation, I wanted to come back as the president or the pope or a .400 baseball hitter. But now I want to come back as the bond market. You can intimidate everybody." - James Carville
http://www.npr.org/sections/money/2016/03/11/470136949/episo...
A better point might be, why were they lent money in the first place?!?
But when bonds default people start losing trust in fiduciary instruments. And trust is much more difficult to build than pure value.
Fannie/Freddie bonds before the housing crash were a perfect example of this. All the information marketing the bonds came with specific warnings that they were not backed by the US government and didn't carry any "explicit or implied" government guarantee. Of course, this turned out to be bunk, and the bond markets knew it, which is why Fannie and Freddie were able to borrow money at lower rates than everyone else. When push came to shove, everyone knew Fannie and Freddie were too big to fail and the government would have to bail them out. I wouldn't be surprised if history repeats itself.
?! I'm not sure what you mean by that, but value is rather hard to create. Certainly the value in real estate and physical goods. Stock value and services are a bit more ephemeral.
But the reason Keyensian hole-digging makes sense is that labour is a wasting good - you can't store the output of people who would otherwise be unemployed, and having people unemployed does long term damage to productivity.
The only sense in which value is destroyed is if the money ends up in the hands of someone who less productive at invest it. But that can't really be predicted. If value is lost in a default, the value was already destroyed in the past, regardless default choice.
Bankruptcy causes real destruction of value as inventory is sold at liquidation prices and the concentration of knowledge and organisation that makes a firm valuable in the first place evaporates.
We aren't talking about literal hole-digging. There are a ton of infrastructure projects in America (highways, bridges, etc.) that could use the labor in construction.
Unless the government, which has the power to break the agreement, breaks the agreement.
The history of finance is well-woven with the thread of sovereign defaults.
I'm just thinking that by the time a state defaults on bonds the "rich man" may be facing a bundle of problems.
How much would diversification help in those circumstances?
For a while during the 07-09 financial crisis, Coca-Cola bonds were priced by the market as less risky than Treasury bonds.
Large multinational corporations are probably safer than the riskiest countries if that's what you're asking.
You can try to leave a general hedge for unknowns, like never giving anything >99% long term confidence. But actually pricing black swan problems accurately should be impossible, pretty much by definition.
In case of mortgage bonds, the bill payers were bond creators who wanted to get the loans off their books as quickly as possible. The buyers should really know better but they too were largely managers whose salary was paid on % of assets basis with no real downside impact on their own wealth.
Note: making wrong assumptions (explicitly stated) about the future is not fraud. I.e., "I think housing will never go down" is not fraud.
When challenged on this they usually claim the first amendment (i.e. license to lie), although S&P went a step further and downgraded the US in a fit of rage when the SEC started investigating them for ratings fraud in 2011. Ironically interest rates on US treasuries went slightly down afterwards.
Various parties are required to trust that anything rated D or whatever is risky. They aren't required to trust that AAAAAA+ means risk-free.
In real life, this sort of sophistry only makes terms like "risk-free" and "black" useless.
At the time, the US government was actively considering whether to default on US Treasuries. The downgrade wasn't some petty revenge.
Their rationale was, in any case, that "entitlements spending" was too high:
https://www.theguardian.com/business/2011/aug/06/sandp-debt-...
The entire rest of the market (and the other ratings agencies) were just about able to determine that if you have the keys to the cash printer you're actually not in danger of running out of cash and political bickering doesn't translate into risk of political suicide.
When actual credit events look likely it triggers a sharp rise in interest rates. It looks like this:
https://4.bp.blogspot.com/-9UVV9J6AdU0/TzxG9W26TsI/AAAAAAAAC...
That did not happen. Nothing happened. Not even a blip. The market (correctly) assessed this as being the usual partisan political bickering.
Even the ghost of a chance of going down to the wire as to whether a bond payment would be missed by a single day on US obligations, is in itself tremendously disruptive. The economy is structured around U.S. debt as being the safest paper there is. Don't downplay just how dangerous dicking around with this for partisan points was.
As I said before: the markets correctly characterized the debt ceiling negotiations as a bunch of drama queens on both sides of the political divide acting up & a storm in a teacup because neither side is politically suicidal.
Ted Yoho, rep for Florida, http://www.rollingstone.com/politics/news/the-tea-partys-gov...
Trump (who isn't in congress, but had a shot at being president: http://mediamatters.org/research/2016/05/08/media-slam-trump...)
yes, it is nuts, but people who vote in congress supported it, and the pres. candidate of one of our two parties also said that.
A government uses debt to pay for stuff. If it defaults on its own debt, it hurts its ability to generate funds without printing currency. So if a government defaults, it not only incites domestic panic and foreign wariness of its markets, it also loses the ability to affordably borrow to pay for programs that can right the ship.
Even for companies where bonds are pricing in a substantial probability of default, it's a big deal if they fail to service their debts. Sprint can survive a 30% drop in its stock price. It probably can't survive defaulting on 30% of its debt.
I fully expect that energy companies have already started to diversify and buy into industries that could disrupt their earnings. Good successful corporations know when to move, RJ Reynolds is probably the best example of a company diversifying when its main product started to become a liability.
Private investors and tax payers are the only victims. The first because their investment is gone and the second who has to pay to fix the the public employee pension funds that may have purchased the same bonds which means the first party is actually stung twice
That's a crass (although depressingly common) straw man of one of Keynes' views.
He mentioned that paying people to dig a hole/fill it up was one way - though emphatically not the most efficient way you could escape from a liquidity trap.
If one looks at the composition of debt 1920-1950 what happened during WWII is the US government essentially took all of the commercial and consumer debt onto it's own books.
"It is curious how common sense, wriggling for an escape from absurd conclusions, has been apt to reach a preference for wholly ‘wasteful’ forms of loan expenditure rather than for partly wasteful forms, which, because they are not wholly wasteful, tend to be judged on strict ‘business’ principles. For example, unemployment relief financed by loans is more readily accepted than the financing of improvements at a charge below the current rate of interest; whilst the form of digging holes in the ground known as gold-mining, which not only adds nothing whatever to the real wealth of the world but involves the disutility of labour, is the most acceptable of all solutions." - Keynes
But do recall that people said the same thing before World War I. All you need is some yahoo to kill the wrong person .
No that is actually not true. You'll find the broken window theory espoused by dingbat modern economists[1] and business people[2] but not hoary old Keynesians and NeoKeynesians. Keynesians believe somethings different which is that if you have a demand and production shortfall due to excess debt and fear (dread zero bound), then the government can help by creating demand. AKA borrow money to pay for labor and materials and use that to build useful stuff. Experience from WWII in the US shows that it'll work even if the stuff is not very useful.
[1] Because modern economics isn't an applied science practitioners are allowed to believe what they want about the real economy.
[2] Leaned about the broken window theory at the country club bar.
Maybe I over-read it at the time, but I got the same feeling after reading a history of the 1500s Mediterranean countries (Braudel's "The Mediterranean and the Mediterranean World in the Age of Philip II"), more exactly that the a very great injustice had been made to the bankers of the time (mostly the Genoese and Fugger family) for not getting their money back from the King of Spain, to which they had lent a lot of money (it's like they had gotten a big haircut on present-day Government bonds). There's an interesting discussion about that at this link: http://history.stackexchange.com/questions/4432/what-were-th...
Breaking windows is a stupid way to crate liquidity though. Better to do some debt-financed spending without destroying anything. For example, the USA's CCC and TVA, that started ending the Great Depression.
Can confirm.
>My rough definition of a financial crisis is that it's when someone borrows money from someone else, and can't pay it back, and it is politically and socially unacceptable not to pay it back.
Nope.
The US Federal Reserve did three rounds of QE, and round three ended in October, 2014. This is not an ongoing program.
https://en.wikipedia.org/wiki/Quantitative_easing#US_QE1.2C_...
Their reasoning is that the bond market is like a giant, distributed bank. Bondholders borrow short-term, and invest long-term, thus providing liquidity for businesses that need to make long term investment.
The problem is, we have no idea what the actual economics of liquidity provision is. The standard Diamond–Dybvig model is used to justify bailing out the banks (and also the "shadow" banks, i.e. the bond market) when things go bad. But if the Diamond–Dybvig model is correct, then liquidity provision is a mechanical process that could be done just as well by the government, and the bond market is just a way to get the free market to do this at exorbitant cost.
We need better theories to understand liquidity. I think the best work in this area is by Holmstrom and Tirole. Holmstrom's nobel price should help raise the profile of this work.
If a riskier company can't renew it's credit, and in general the money is getting more expensive won't this lead to bankruptcy for many energy companies. Those companies won't be able to pay their debt back so ergo the money has to be written off.
This would lead to higher IR on any old-energy based debt which could then put even the less risky companies at risk as they need to put more of their profits into servicing their debt.
There would be a systemic failure. Unlike the housing crisis it will be due to companies that can't afford their debt in the future, rather than overextended home owners who couldn't afford the debt from day zero. But the effect could be the same.
Disclaimer: Armchair economics.
I'll give you another reason - It's Bloomberg News. One of the Buzzfeeds of the financial press.
This argument is generally pursuasive to me most of the time. Should I be convinced otherwise in the present case. -a layman.
But also, even if there was vastly more oil than thought, that wouldn't matter because if you dig it up and burn it, Earth turns into Venus. Peak "oil we can afford to dig out without fucking ourselves over thoroughly" is already in the past and receding.
If just one of the big banks were to fail it could take down the entire economy along with it. Even after all the stress tests... Because they don't count the interconnectedness of these backs.
For this reason even the Libertarian and staunch capitalists should be clamoring to break them up into smaller chunks. There's no reason why such an action should be bad for investors or the banks themselves. It could merely be a forced diversification.
If government intervention can prevent a lot of people falling into poverty/homelessness at the cost of slightly delaying technological progress, I think that can be a price worth paying.
So in a more general sense, why not simply pay the workers of an obsolete industry? I'm not sure I have a good answer to that, since I'm not educated in economics. But I would guess that a dying industry can still generate net wealth, just not enough to sustain itself. It depends on how much of its costs are labor vs infrastructure.
An industry that cannot generate enough wealth to sustain its infrastructure is probably generating negative net wealth, and it's better policy to let it die and give money directly to the workers.
But if an industry can sustain its infrastructure without government help, but it can no longer afford the labor costs, I think the industry can still generate positive net wealth for society if the government plugs that gap. The portion of the money the industry is paying out of its own pocket to its workers is already net gain for society, why throw that away?
Protectionism has its place, but we should fully evaluate the cost, damage and benefits created by protectionist policies before implementing them, otherwise you end up with massively subsidized US corn crushing our NAFTA partner's agriculture industry, and Chinese Steel being dumped on the US at significantly lower prices than the US Steel industry can produce said metal at (mostly due to Chinese state subsidies, just like our agribusiness subsidies).
Note that I did not claim that protectionism should be the default policy. My only claim is: it would be silly for a government to refuse to consider protectionist solutions to certain problems, simply because capitalist dogma says it shouldn't.
Anyway, batteries are not an energy source, just storage. Somebody has to build generating capacity.
IIUC, you can generate far more solar electricity during the day than can be consumed. But that electricity isn't available at night because storing it in batteries is too expensive right now. With cheaper batteries there are stronger incentives to build more renewable generating capacity, driving down costs further etc etc.
Back in the 1960s and 1970s when the power industry thought we were going to go nuclear, they built a lot of pumped water storage systems. There's one in Western MA built under a mountain that pumps water from a river to an artificial lake at the top of the mountain. The plan was to pump water at night using extra power generated by nuclear plants, and then run generators during the day to make up the shortcoming.
My guess is that the facility now pumps water during the day to act as storage for solar power.
The efficiency is about 70 or 80%, from what I remember. Batteries are a much higher efficiency, and can be placed much closer to the load.
Thus, mass-produced low-cost batteries will really disrupt the grid, because older forms of storage, like pumped water, will be obsolete. Furthermore, because batteries can be placed much closer to load, the needs for grid capacity change because high transmission lines no longer need to handle peak load.
Someone could take the old Superconducting Supercollider tunnel in Texas, put in the theoretical PowerLoop infrastructure for about $500m and then it would be able to store enough power for the entire state of Texas to operate for over a day. The money it would make on the arbitrage between cheap power generated from solar & wind and times where demand exceeded baseline generation capacities would pay for the system, according to the designer, within a few weeks.
Not explained in the article is there is a threshold effect in play here. Electric cars are probably cheaper on a per mile basis now, but not on a capital basis. However batteries are coming down in price so fast that within a decade we may find that electric cars are cheaper on a capital basis. If that happens, gasoline cars will stop being produced overnight.
Other one that people don't consider as much is that outside the united states and a few other oil producing countries oil imports are considered to be a necessary evil. If the governments of those places think that they can run their transportation infrastructure without oil, they'll switch as fast as possible and erect very stiff taxes to reduce demand for oil imports. AKA do the rulers of China like that they need to import oil? No they absolutely hate that. Which is to say inside the US oil companies have a lot of economic and political power, outside the US governments hate and fear them.
Both of these means that demand for oil could start falling rapidly. That will depress the price, any producer that can't economically produce at the new price will be bankrupt. Any oil exploration business will go bankrupts.
Some people will still pay for range or range at speed or less charging time.
http://jalopnik.com/they-drove-a-tesla-from-la-to-new-york-i... is a decent look at the state of technology. Driving across the US takes at least 12.5 hours of charging time, and averaged 65.4mph while ICE record averaged 98mph.
Even when "electric cars are cheaper on a capital basis", the technology will still be well behind ICE in some ways (and well ahead in others).
% Type mpg gal/yr ext
50% Electric -- 0 0
25% Plug in Hybrid 80 150 X 0.25 = 37.5
25% Non-Hybrid 30 400 X 0.25 = 100
-----------------------------------------
Total 137.5
100% Non-Hybrid 30 400 X 1.00 = 400
------------------------------------------
100% 400
And that's in the US. In countries without domestic oil production the choice will be made for you.
I have seen far too many "projections" that assume nothing will change - they are always wrong, and thus worthless (unless it's your job to make the change).
Because the hype will be with the new, and auto makers will be perceived as "old" even if they release electric cars, which will be tough because they're already behind Tesla.
Everyone wants to have that Tesla, and no one wants a Blackberry even if the newer versions have touch and have all the features of an iPhone.
It happened in the smartphone industry, with taxi apps etc. etc. It's hard to shake off widespread public perception short of reinventing yourself which is possible, but quite rarely seen.
I'm not even sure this is the case. Tesla, with AP2.0, seems to be starting from scratch. New cars will have no autonomous features. I'm not convinced they have a technological lead over other players in what is becoming a saturated space.
More broadly I think you are right, it's hard to see where the different groups really stand from the outside.
I agree, this was a pretty slick move all around. Add the hardware (which is simple and cheap), collect data, iterate on the technology, and sell it now as a "game-changer" and "upgrade". Elon has a knack for packaging things like this.
The car manufacturers don't appear to be ignoring tesla at all. In fact, Tesla is being outsold by more than one of the "old" companies electric cars today.
If you have to compare it to phones, it's much more similar to iphone vs android. Android might not have been first, but they responded quickly.. a variety of phones/cars allows manufacturers to meet the requirements of more market segments.
At the end of the day, Tesla/Apple can only offer a handful of models.. if you don't like it, you're choices are the large variety produced by Android manufacturers. That virtually guarantees Android will have a greater market share than the iphone/Tesla.
Powerwall clearly is not a complimentary product of a car. If it was wildly successful, it would have had minimal effect on Tesla's car sales, while having a significant effect on it's battery production.
Similarly, SolarCity (which is merging with Tesla) increases demand for batteries.. but again, has little effect on their car sales.
In fact, the one thing all of their products share in common (solar, powerwall, cars) is that they increase the demand for batteries.
Tesla may be a car manufacturer.. but it would easy to think their primary business is actually batteries.
And, while Tesla cars might be bragworthy and worthy of showing off to your friends, house batteries are not. All that really matters is cost and efficiency.
It's like storage, only it costs almost nothing: all you need is a lot of cars with overly-large batteries plugged in, and a little predictive software.
Also, existing EV owners don't have an EV <> Household bidirectional interface connected, and I'm not aware of any serious vehicle manufacturers claiming they will do this. Probably because if the battery has an X kilometer / Y year warranty, how does using it as a household power reserve affect that warranty?
An X Km warranty is just an X KWHr warranty in different units.
Here in Japan, Nissan has been advertising the capability on TV, and the equipment is available for general sale
> no one wants a Blackberry even if the newer versions have ...
I agree that there is a hype aspect to this, but I don't think it's the whole story. You rightly point out that it's hard to shake off widespread public perception short of reinventing yourself. I think that in many cases the perception is earned.
As is often pointed out, there are many aspects of the iPhone that are not original to Apple. Tesla isn't the first company to come out with an electric car. I don't think hype alone is going to have people shelling out tens of thousands of dollars for a new car.
What prevented the established companies from putting out compelling products themselves? After all, they already have experience in the industry. Clayton M. Christensen's "The Innovator's Dilemma" discusses this in part. And once you've lost the perception of being competitive, you're playing catch up on both marketing and product. Ouch.
The Honda Insight hit big Western markets before the Toyota Prius around the turn of the century, but ultimately the Prius became the most popular hybrid vehicle.
What customers want in a car is different from what they want in a phone.
The (original) Honda Insight was a tiny 2-seater of a car, more tech demo than "ordinary" car, and perhaps it's the gift of hindsight but the Prius' popularity for with people that own one car, as a 4-door 5-seater car vs a non-sportscar 2-door 2-seater car seems all but guaranteed. (How many people do you know that own 2-seaters vs 5-seaters?)
Similarly, the vast majority of smartphones these days are (unfortunately) glass bricks, as first popularized by Apple. There were other interesting designs out there, but ultimately the iPhone design became the most popular smartphone design.
...in Silicon Valley. Tesla is far smaller than you think. The market and mind share outside of extremely wealthy sectors of the country is miniscule. Nobody where I live has even heard of them (unless, of course, you are a technically inclined forum reader).
HN seems to think Tesla has already won (even though they can't profitably produce a vehicle), and that everyone working for the auto majors are old, stodgy, and stupid. But I'm betting otherwise. I think they see that the writing is on the wall, and they're allowing Tesla to do much of the leg work.
No, this hasn't been "debunked".
Have you ever looked at what all the auto majors spend on R&D? Do you you think that it's $0? Both GM and Ford spent $7 Billion on R&D last year. Tesla spent under $1BB. Yes, those old, stodgy, do-nothing companies outspent Tesla by a factor of 10-1 on R&D. This is a heavy R&D based industry.
If an EV supports 120V outlets, then chargers are already more common than gas stations by many orders of magnitude. Fast charging is a different story... But for long road trips, taking a break for a couple hours every 400 miles doesn't sound like a bad idea, and we already have ample rest stops along most highways.
People like to make it sound like cultural problems are insurmountable. Sometimes, that may be true. Other times... we just have to suck it up and realize: The future is already here, and it requires change. Personally... I'm excited to get away from a gasoline automotive economy -- at least electricity is agnostic as to the source of EMF.
(In fact, a whole bunch of car batteries hooked to the grid sounds a lot like distributed energy storage required to smooth demand curves...)
I mean, doing 1500km travel (which i do 1-2x per year) would mean 4 hour drive, then couple of hours of forced wait, again drive and so on? completely useless
If you regularly drive more than 300km, then yes, ICE vehicles will remain your only option. But if you are a suburbanite or city-dweller who has a typical round trip commute of 70 miles or less[1] then in the very near future a Tesla Model 3 or equivalent will be the economically rational option for you to purchase.
I think about buying a EV in the same way as I would think about choosing the number of bedrooms in a hypothetical house purchase: do I overpay for a house with one extra bedroom, in the rare case I have relatives or friends once or twice a year, or do I offer to put them up in a nearby fancy hotel for that rare case? Similarly, do I buy the car that fits my needs for 95% of my day to day driving, is economical to run, and rent a car when I need the extra range, or do I overpay day to day for the rare edge case where I need that extra range?
Then again, plenty of people choose ridiculously oversized SUVs to drive their kids to school so we shouldn't expect drivers to suddenly become economically rational in their vehicle choice.
> I mean, doing 1500km travel (which i do 1-2x per year)
which is it? Things you do once or twice per year are exactly what the GP is talking about. Rent a car that can.
Besides, I believe the Model S can go for about 400km at 130 km/h, so realistically you'd have two half-hour breaks at a supercharger. It's not so bad if you combine it with lunch / coffee. If you were to take the recommended driving breaks (at least 15 minutes every 2 hours) and increase the interval from 2 to 3 hours (because people who do 1500km pushes probably don't take as many breaks as they should), then you end up with 30 minutes more. That's not ideal, but not too bad either.
And then you don't need to go out of your way to buy gas!
But, that considered: I see a lot of stories about developments in rapid charging. IMO, in about 5-10 years an EV will be able to charge to 90% capacity in 10 minutes.
That's about as long as I spend at a gas station when I drive long distances. (I usually go to the bathroom, stretch my legs, ect.)
that's assuming people have more than one car. a person can get a rental car, this is added mental task that you're asking a person to do. if you're requiring the user to think/work you're going to lose a lot of users.
That's the marketing angle that will be played, obviously for people here in a general socioeconomic sense there's no problem affording an EV and no problem getting a rental at the drop of a hat anywhere on the planet. Like claiming those airplane thingies will never catch on because people don't like renting cars and we are all familiar with suddenly having a transoceanic or cross continent travel dropped in our laps so we'll all be forced to live on houseboats or something. I mean, everybody travels all the time, so no body should be permitted to own a car that isn't for travelers, right?
The marketing push we can all look forward to is exhaustive explanation via legacy TV commercials and perhaps product placement on sitcoms, etc, that only poor people can't afford to rent a car for weird long trips and being poor isn't cool, so that'll kinda take care of itself.
Unless you forget to plug the car in the night before.
Unlike switching the lights off you have to get out of the car to plug it in.
If the car is going to beep before you get out then it's not a reminder for something I've forgotten,as with lights left on, it's just irritating UX noise, like the "don't forget your mobile phone" message. After a very short time I go blind to that.
If it's an alarm on a mobile/smart watch it's still a tedious chore to have to plug in a car when it's raining or I'm in a rush or I've parked in a spot without a charge point.
People _will_ forget to charge their cars, just as they forget to charge their phones and they will be late because of it.
If purely electric cars are to be brilliantly convenient everyday repacements for fossil fuels they'd be better with either some kind of induction charging, fast charging that doesn't take hours or hydrogen fuel cells.
If we had better ways of splitting hydrogen we could "fast charge" anywhere water and electricity is available.
But no, it's not a huge issue. New cars seem to come with apps nowadays, so as you said your car can just send you a notification to your phone if you haven't plugged it in five minutes after arriving back home. And maybe another one if it's still not plugged in before you go to bed.
Unavailability of charging (see sibling comment) is a bigger issue that won't be as easy to tackle. But reminding users to charge their car isn't what's hindering EV adoption. You can be late if you forget to fill up your petrol car, too. That can easily take ten minutes if there's a bit of a queue.
Unless, like many people in my city, you park on the roadside because your 19th and early 20th century houses don't have carparks.
Having an EV-only inductive charging parking spot every here and there might help in the future (https://en.wikipedia.org/wiki/Inductive_charging#Electric_ve... - it's coming), but probably will neither be cheap to retrofit nor very practical.
This is an issue that needs to be solved for EVs to become practical here.
Much like how Red Bull completely replaced Coca-Cola.
Or that time Microsoft was replaced on the desktop by Ubuntu.
Or that occasion when the Segway killed off bicycles.
Sometimes the incumbent wins. Sometimes they don't.
Making sweeping generalisations across all competitive landscapes in all industries isn't a very effective method of forecasting in particular industries.
For most of the history of the automobile, the UX was standard and very simple. Now there's a major expectation-resetting going on. The UX of a Tesla is not something that just can be bought from a third-party supplier and bolted on like a fuel injector or axle or radio. At best, the software departments of big automakers are accustomed to building embedded systems. Making the transition to will not be easy.
https://www.youtube.com/watch?v=HM1-Yh5fwB8
Building a good embedded touchscreen UI is not so much rocket science as it was just 5-7 years ago. There are a lot more people with the knowledge nowadays - because of the mobile app revolution.
The iphone appeared as soon as the hardware shrunk enough to be able to support it. Earlier tablets failed mainly because the hardware just wasn't capable enough.
The car itself is a complicated platform with many components. How they communicate will affect how easy it is for the developers to build reliable, secure, components. (e.g. braking systems preferably shouldn't be open to everyone https://www.wired.com/2015/07/hackers-remotely-kill-jeep-hig...).
Then you have the cloud aspect, which allows users to access car data and functionality from apps, allows the car manufacturer to do analytics, etc. Now you have a car company that's responsible for running safety-critical servers in the cloud. Again, this leads to huge variation in reliability, developer experience, and security (e.g. your car's climate control is best left to yourself and not everyone else https://www.troyhunt.com/controlling-vehicle-features-of-nis...).
TLDR Software is eating cars.
I want tactile controls (knobs and levers and pushbuttons) I can use by feel without taking my eyes off the road.
And, of course the UX of a Tesla is something that just can be bought from a third-party supplier and bolted on like a fuel injector or axle or radio. It's just a matter of hiring some smart developers. Many readers here could probably do it in a weekend ;-).
to adjust heating/volume/switch song/etc i don't need to lose focus on road, i just know which knob to turn or which button to press. I know by experience that when i look at central panel, nuke can blow in front of my car and my eyes would barely register it since it's not in my direct focus - very, very dangerous.
it feels much more premium to have dedicated physical controls for all required tasks. fuck tablets in cars, it's horrible idea that should be put only into cheapest cars, or straight forbidden.
Energy transitions from wood to coal to natural gas to renewables etc... usually take decades to switch over.
As well, there's no exponential growth in solar. We may be hitting the S-curve flat line on battery storage.
We can't keep moving left on the periodic table. Lead Acid ---> Lithium Ion. Where to move to next?
Lithium Fluoride batteries are scary because of the reactivity...got that noble gas, ya know?
Zinc ion rumors abound, but those ions are big, really tough. Zinc is a huge ion and thus you can only stick a few in the electrode.
Any chemists on hackernews or am I forever alone?
As for the shorter term, have you looked at SolidEnergy? If their tech lives up to their promises I think it could really tip the scales for EVs. If it's relevant for grid storage I don't know.
[1] https://news.cnrs.fr/articles/a-battery-revolution-in-motion
[Side note: some interesting quotes from the above link... In 2012, the French researcher decided to take the bull by the horns and make the most of France's know-how in sodium batteries. "For lithium, all fundamental research had been conducted in Europe, especially in France," Tarascon points out. "Yet it was in Japan that the technology was transferred and brought to market, allowing Sony to launch its first lithium-ion battery in 1991. As a result, 95% of Li-ion production today takes place in Asia." It is out of the question to repeat history. The CNRS (responsible for fundamental research) and the LITEN-CEA (in charge of technology transfer) have thus joined forces with around 15 industrial players such as Renault, Saft, and Alstom to create the RS2E network dedicated to new-generation batteries. Their stated objective is to ensure research AND development, in order to bring sodium-ion batteries to market on European soil as soon as they are ready.]
For stationary energy storage energy density is even less important and costs are even more important. IMO the best way to make stationary energy storage affordable would be to increase battery cycle life, via chemistry tweaks and/or charge controller tweaks. You don't have to compete so fiercely on lowering up-front costs if you're selling to institutional customers and can demonstrate lowered lifetime costs.
I'm not going to proclaim that lithium ion batteries will certainly win the storage wars. But I think there's an interesting historical example from photovoltaics. In 1976 terrestrial PV cells were mostly made out of crystalline silicon and a lot of researchers were searching for materials that could lower PV costs either by increasing conversion efficiency or by being cheaper than high purity silicon. The search never stopped, and maybe some day silicon will be overthrown. But in 2016 crystalline silicon is still the most common, most efficient, and least expensive option for making terrestrial PV cells. Iterative improvements on the incumbent technology kept it ahead of all the many disruptive alternatives proposed over the decades. Maybe the same will happen with battery chemistry (either lithium ion or another one that's already been commercialized, like sodium ion) -- iterative improvements that keep disruptors forever playing catch-up.
The article is a (lousy) summary of a report from Fitch. It has no new content, and its regurgitation of the report confuses most of the fundamentals of the energy and credit industries.
I agree that the article was very click baity, but I am curious about how energy utilities will be affected by cheap batteries. These are some of the most entrenched and connected industries, so disruption, if it happens, won't be easy or fast.
Most egregiously, Bloomberg is cherry-picking the words "death spiral" out of somewhere in the middle of the report and putting it in the headline.
A few things (like shale oil drilling investments) might be impacted in a "death spiral" way, but I don't believe that is anywhere near the 25% they claim by invoking everything that touches oil.
Also, bonds have a finite lifetime and are often for equipment with depreciation. Even if 25% of bonds today were for oil infrastructure, many would be re-paid before the disruption had significant financial implications.
TLDR: the headline is egregious clickbait of a regurgitated article.
I enjoy the comments on places like HN, because it provides a good accompaniment to Wikipedia: where Wikipedia is a summary of the topic, as it stands, comments on HN are a summary of the discussion about the topic, as it stands. (Well, when we get at least two different perspectives contributing and discussing, which is why we should be careful with groupthink and downvoting dissenting views.)
In a vertical market like cars, you've got many actors, from iron ore mining to the "going from A to B reliably" service, going through part makers (who actually make most of the car), car maker (brands), repairmen, dealers, etc. Very often in those markets, one actor controls a key level, protected by entry costs, secrets, regulations, exclusive customer relationship, whatever. By leveraging this stronghold, this actor extracts most of the margins, leaving just enough to other levels to let them survive. You want all other levels in your vertical market to be commoditised: Microsoft wants PC hardware commoditised, Über wants driving to be commoditised (through underpaid drivers first, unpaid robots ASAP), etc.
It seems that for cars, mastering the engine is the way to control the whole vertical market. Moreover, ever-more-stringent emission + fuel efficiency regulations keep it hard to make engines, despite technical progress. Electrical cars change that, because it's (comparatively) very easy to make an electrical engine.
Another asset held by carmaker is the brand and consumer relationship, and this one's probably doomed as well. Cars stay parked for 95%+ of their lives, the only reason you buy one (and it's your most expensive possession after your house) is because you want it to be reliably available whenever you want. Once cars drive themselves, it stops making economical sense to own one: for a fraction of the price, you'll rather subscribe to a fleet of autonomous cars which reliably deliver one on your front door 10 minutes after you summoned it with your smartphone. That not only means fewer cars driving a higher percentage of time: it makes obsolete the car-makers' genuine expertise in making individuals want to buy their models. I'd bet that selling hundreds of thousands of units to Kalanick is significantly tougher than making some hillbilly lust for a Canyonero.
It looks like mastering level-4 autonomy will be the new way to lock that vertical market, and current car-makers don't look better positioned than the Silicon Valley to secure that stronghold. Cue the famous last words by the then Palm CEO, saying about the iPhone “PC guys are not going to just figure this out. They’re not going to just walk in”. Yes, in radical changes, it typically is an incumbent who figures it out.
Interesting, I'd picked a different idea as the market shape ("AirBNB for self-driving cars").
Yours probably works better for a well-capitalised firm. Especially since it encourages lockin through mere inertia.
There's no reason the manufacturers can't take this approach themselves. With a self-driving fleet, they can bypass their dealer channels and undercut everyone else, especially Uber, and Lyft (less Tesla).
In short, it's the GE model for aircraft engine. Selling engine hours, not engines. You can charge a premium over the actual financial costs because it's simple.
Plus most buyers prefer a fixed, regular payment over paying-per-use, even if doing so is financially irrational. Loss aversion and whatnot.
AirBnB works great when you exceptionally need accommodation, far from your normal environment. Most people would hate to need booking an AirBnB every night of the year, in order not to sleep under a bridge.
I think people can regularly move around in impersonal vehicles (as shown by public transportations), but they'd hate depending daily on borrowing a car that's clearly someone else's. They'd probably also have a problem leasing their own car when there's a slight risk that they might need it. So really, the dynamics that makes AirBnB possible doesn't seem to translate well to daily transportation.
> There's no reason the manufacturers can't take this approach themselves.
I'd turn it the other way: it's likely that the fleet operators will also build and repair the vehicles. But I wouldn't bet that legacy car manufacturers will be the ones cornering that market.
Very unluckily for them, electric power and autonomous driving look like they'll become mainstream simultaneously. As explained above, I believe the former will breach their historical moat, and nothing indicates that they'll be the first to rock at the latter, as well as the new kind of consumer relationship that comes with fleet operating for consumers (most of them already operate corporate fleets).
Anyway, unless you're a shareholder, it doesn't matter much whether the behemoth of tomorrow is called Ford, Tesla or Über. Many of the individual engineers will be the same either way, hired by the winner. I don't believe in Apple winning this: they're good at making you lust for ownership of tangible goods, but offering collective services in impersonal vehicles doesn't feel like it's in their DNA. As for Google's moonshots, I can't remember any of them being successful. Moreover, they'll probably be beaten at their usual strategy of accumulating more data than others and parsing it better: Tesla's already accumulating millions of analysable miles with their actual fleet of actual cars doing actual travels, while Google toys around SF with replicas of Noddy's car. And Musk's been hinting at interest for building a company-owned fleet (although he would be a fool to say it out loud: Tesla's all about making electric cars sexy to own).
Über is also very well positioned for smart data accumulation. And IIRC, Kalanick has always been open about his plan of cornering the autonomous cars fleet market before autonomous cars exist, and hiring drivers as a temporary stop-gap (which he also openly treats like shit).
> Plus most buyers prefer a fixed, regular payment over paying-per-use
They currently pay a mix of the two (they pay for gas and maintenance per mile). With electricity being cheaper than oil, it's possible for fleet operators to offer an even better per-month / per-mile ratio. And I agree with you that they'll probably do just that. Unless of course, a monopoly is secured so fast that there's no point in making fleet customers happier than those of cellphone or cable operators.
Some people use Uber or Lyft occasionally. Some use them every day. In both cases the mechanics of use and billing are identical.
> I'd turn it the other way: it's likely that the fleet operators will also build and repair the vehicles.
Hertz, Avis et al are not seen as competitors to GM or Toyota, because the car-hours market is relatively dominated by owner-operators.
I don't see anyone talking about those companies manufacturing their own cars. Probably because manufacturing a modern car is a hard business to get started in. Tesla did, after years and many millions of dollars.
Maybe Tesla will go the fleet model. But nothing stops Chevy from doing the same. Nothing stops every single manufacturer from doing the same. There is no network-effect moat here.
They aren't: you have to go and fetch the car at their shop, which is very impractical compared to having your car parked on your driveway.
The most radical change with autonomous cars isn't that they drive themselves with passengers in it: it's that they'll drive themselves empty in order to pick up passenger wherever they want, whenever they want. That's what makes not owning a car practical.
> There is no network-effect moat here.
There might not be a natural monopoly, but there is a pretty big critical mass: to reliably serve cars fast enough at predictable prices, a fleet must be pretty large. And whenever you go from city A to city B, you need the operator to be present in both cities.
You're right that everyone's allowed to compete on this new market, including Chevy. Just as Blackberry and Nokia and Microsoft were allowed to compete in the post-iPhone smartphone market. But have they got a critical advantage to help them succeed? They've got plenty of legacy that will be more of an hindrance: dealership networks (that's another thing Tesla is fighting tooth and nail not to get entrapped with), many engineers and engineers-turned-executive who can't let go of a dead culture (ICE, driving as a dangerous-but-exhilarating activity, cars marketed as phallic substitutes), syndicated workers on their assembly chains...
Twice in my career I've seen incumbents lose their market, because it took them too long to embrace inevitable change: Unix workstations killed by Linux on beefed-up PC in the early 00's, and cell phone makers deprecated by iPhone and Android in the late 00's. I'd be surprised if GM and Ford proved smarter and more adaptable that HP or Sun in the good old days, or Nokia and RIM a decade later.
When the paradigm shifts, you're better off with a clean slate than with a legacy.
Precisely. Some people will still want to own their car, even if it's autonomous. I imagine that over time folk will become accustomed to living more like New Yorkers -- renting car time -- but it might not be a big bang.
> And whenever you go from city A to city B, you need the operator to be present in both cities.
Or you have a service acting as middleman. Like AirBNB, or Amazon, or Google.
> Unix workstations killed by Linux on beefed-up PC in the early 00's
This process has already played out in the car market. It's what Ford did to the dozens of small manufacturers who existed previously. Unix vendors had high margins that couldn't be sustained as mass-market technology caught up. Sheer manufacturing power (and let's not forget Windows here) did them in.
New car manufacturers are not entering a market of Unices. They are entering a market of Intels.
> cell phone makers deprecated by iPhone and Android in the late 00's
This one is due to genuine product improvement, in my view. The iPhone was actually usable, it wasn't just random options stuffed any old where the engineers found a gap in the menus.
But again, cars are a rather more sophisticated product. They already compete on design. There might be some more competition on usability.
> I'd be surprised if GM and Ford proved smarter and more adaptable
It occurs to me that I ought to have disclosed that I work for Pivotal, in which Ford recently invested a lot of money. It didn't occur to me because I don't work with any Ford teams myself, though some colleagues in the company do.
What I can see is that neither Ford, nor GM, are just sitting around waiting to be replaced. The lessons of the technology industry are now widely diffused. Applying them uncritically means assuming incumbents will rest on their laurels, because that's what happened in tech.
Except everyone else has gotten to observe what happened in tech. Incumbents understand the perils of laurel-resting.
> When the paradigm shifts, you're better off with a clean slate than with a legacy.
iOS is based on a technology that is nearly 50 years old. Android is based on a technology that's 25, inspired by the technology that is nearly 50.
<Very often in those markets, one actor controls a key level, protected by..
Any writings on who these actors in different markets? clothing? retail? shipping? whatever?
<Once cars drive themselves, it stops making economical sense to own one
Driving is so much fun. Do you ever think they will have laws reducing the driver-cars? Like HOV but for NDV?
So.. when you get into a house you no longer need a landline 'cause of cellular phones. Does this mean when you get into a house you no longer need to own a car? Do you really think this will happen?
Another observation. reporting of deaths.
Now, whenever a passenger plane crashes, it becomes top of search news. "We Interrupt this broadcast" historical procedures. Same for bad rail accidents. Yet with automobile accidents, there are over 30k each year on our highways. Not "interrupt this broadcast" or top of search news. [some, but the weather chair reactions or extreme pileups]. It is even hard to find news about the death near your home highway at times.
But, anytime a driverless car is involved in an accident, especially a death, we can expect them to be top of search results and lead news story. It is new. It will be years before an accident with a driverless car becomes so routine it does not make news. Tons of news sites need fodder. A driverless accident is perfect food for clicks and genuine interest.
<deliver one on your front door 10 minutes after you summoned it with your smartphone
I thought about this. Okay, I will not trust anybody to be better than me when I drive in a snow storm. Several months I HAVE to go to work. I really doubt I will take a driverless car especially because I know how crazy other drivers are and I know the idiosyncrisies of the trek I take 530 times a year. Thus, I will have to own a car. It will make economic sense.
So is horse-riding, and some people still own personal horses. But most people don't, and those who do ride for fun, not for transportation. They take their car to go and ride their horse actually. Because horses are comparatively impractical.
When autonomous cars will be mainstream enough, there won't be any traffic jams in autonomous-only lanes (yes of course it will be a thing), contrary to those allowed to legacy cars. Also, fleet cars won't park, so the pressure to reclaim parking space in the cities will become enormous, and parking your personal car will become as hard as feeding your horses while shopping at Walmart. Human-driven cars will become as impractical compared to fleet cars as horses are compared to cars today.
> reporting of deaths
It's hard to predict which deaths remain newsworthy, and I don't feel like I can. However, I suspect that crashes between two autonomous cars will be very rare, compared to human driven vs. autonomous car crashes. And the primate-operated vehicle will be blamed, probably rightly. Musk considers he needs autonomous cars to be 10x safer than legacy ones, in order to overcome this effect, and expects to reach that level in a couple of years (he wrote that's when he'll remove the "beta" qualifier on his auto-pilot).
> when I drive in a snow storm.
Tesla's radars see perfectly well in the worse snowstorms, you don't. The car has reflexes orders of magnitude faster than yours, a direct connection to ABS sensors, etc. As of today, Tesla cars monitor their human drivers in every driving conditions, compares it to what the autopilot would have done, and whenever a discrepancy occurs, it sends a report over GSM to Tesla's engineering teams. That's how they improve their autopilot. As soon as you drive one you're teaching every car to drive better than you in pathological conditions :-)
Don't know if it is intentional.
Why did you ever expected those to not happen at the same time? It's expensive oil that makes people look for alternatives.
The left was right about peak oil. Just for the wrong reasons. People will stop using oil, not because we ran out, but because the alternatives will just be so much better.
I don't think people really understand how much oil is out there. We've got more oil in the Colorado shale formations, than Saudia Arabia does! It is fairly expensive to extract, but its existence means that oil prices will never 'consistently' stay above 100-120$ a barrel for more than a couple years. Nightmare 200$/barrel "peak oil" are just never going to happen.
Wonder what city councils are going to do to supplement the revenue lost if and when income from parking spot use shrinks due to a combination of smaller fleets and individual cars in the fleet spending less time parked up.
Of the 4 exponentially improving technology clusters mentioned: battery storage, electric vehicles, self-driving hw/sw, solar pv -- I'm trying to imagine if any one of them might not ramp up within the 5 to 10 year time scale Tony is talking about. And I guess it would have to be the software[1] part of the self-driving hardware (sensors + compute) / software combo. And perhaps lithium[2] ore extraction. I don't know enough about the technologies of solar pv to hazard a guess. If battery storage happens we definitely get electric vehicles (along with better laptops, tablets, and phones).
[1] https://www.udacity.com/course/self-driving-car-engineer-nan...
1. Battery cost is falling and EVs will be as affordable as their gasoline counterparts in six years
2. It will take until 2030-2040 until the demand for oil starts contracting
It seems to me like you get to pick only one of these statements. Once EVs reach affordability parity with ICE autos my guess is that things will hit the steep part of an S-curve.
And some ICE cars would also be sold in non-negligible amounts if those cars merely reached parity. You actually want them to be more affordable than the competition.
But factor in cheaper EV cost of driving, and posit:
taxis, delivery vehicles - changes fast to cheapest (EV)
Private and corporate leases - 2-3 years between replacements. Will move to cheapest option (EV)
How much of total driving is made in these categories and how much is "long tail" of private driving in old cars? S curve I say.
And general inertia of change. Think more recently how long people held onto blackberries after the glass front smartphone took off. Few would now argue the merits of the keyboard over the full screen (not trying to trigger those of you left!) but a fair chunk held to the traditional view for many years. The same will be for the above machines. From this the timeline seems reasonable for me.
But I don't think it does. If (when?) you can hail a self-driving cab for a lower per-mile fee than the cost of gas+insurance+car payment, Uber (Tesla?) can (will?) put electrics on the street as fast as they can be financed (manufactured?).
People will let their vehicles rust if it's cheaper to take a cab. No major consumer purchases required, just cultural diffusion at the speed of super bowl ads.
1. Those statements are attributed to different organizations (Bloomberg New Energy Finance and World Energy Council respectively) but are quoted in the same article... They maybe be inconsistent!
2. If you consider integrated (or global) oil demand, you can image a world in which they are both true: EVs may become cost competitive with gas in markets w/high gas + car prices (eg Norway) but not in markets with lower prices (for both goods) where EV takes longer to cross a 'tipping point', delaying the global oil demand "peak" to a later date. There is no one homogeneous "global market" but a bunch of different local markets w/different cross-over dates. (I suspect the 1st article is talked about market for certain products - but not global.)
3. Oil use will also keep growing even as the % of EVs sold starts to increase - b/c of other uses (e.g. trains/heating etc.) that are linked to growing population/gdp.
Newspapers might be the closest parallel, but they haven't completely crashed yet and might well bounce back.
Who will probably provide the template for the decline of petrol: fight like hell in the courts and lobbies to try and stay relevant in the first world, use ISDS mechanisms and so on, while cranking up campaigns in the third world to cover your losses.
I mean, I'm no 'peak oil' person or anything, I just don't trust that we went so quickly from the idea that we were running out of oil and that it would as a result become super expensive now to this idea that it won't be worth pulling oil out of the ground for much longer; I strongly suspect that the truth is between those two scenarios, and would like to lay some money on that suspicion.
(Unlike many people, I don't think I am smarter than the market... But I do think I am less excitable and often have a longer time horizon. And yes, I know that I am probably just as wrong as the people who think they are smarter than the market.)
Think Oil Sands: massive amounts of Oil in Alberta and Venezuela was not considered in reserves just a decade ago because it was not economically feasible to access it. Now it is. Deep Oil drilling, arctic exploration - so much changes so quickly. Heck - most wells are left with a lot of Oil in them because it cannot be extracted.
Of course eventually there will be such a thing as 'peak oil' but it's hard to determine.
As the world economy expands, because in 10 years everyone will be on a smartphone, the demand for plastics and fertilizer and rubber and roads etc. that demand should expand more than fast enough to cover the fleet conversion from gas to electric.
200 million vehicles in the USA fleet, and they sell 17 million in a good year. 12 years to replace the fleet; it is currently about 11 years old on average.
I think the article is over stating it. There won't be any problem, it will take decades to adjust anyway.
Is the y-axis of that graph actually oil demand?
Best case scenario, according to the graphs in the article, is if "Low Carbon Policies" come into effect to disrupt the oil industry as quickly as possible... we will still be consuming 50 million barrels of oil per day in the year 2060.
There's never been a case in the past when that assumption didn't turn out to be true, of course..
So, in 6 years, I can replace my Corolla with a Tesla model 3. Cool!
A less cynical person might says "Those companies" have a strong incentive to preempt and partake in the disruption. And it is my belief that's the path most of them will take.
The big secret in a lot of countries where energy and communications are state monopolies or semi-private corporations is that they are LOADED with debt that the govt used as a private slush fund.
If you are looking for "off balance sheet" liabilities in a government, look at its utilities that are exempt from public records laws. This is where a lot of the bodies are buried.
If people start dropping off the grid, states will just convert the power bill to a straight up tax bill, but without the pretense of a service attached to it. There are already precedents for charging electric vehicle taxes to make up for the shortfall in fuel taxes they create. Govts say, "well, if they are using the roads, they should pay for them." except gas taxes go into the slop bucket of general revenue, and the roads have been paid for through other taxes. The largest state expense is employee salaries, yet all road maintenance is done by contractors, so the "roads" thing doesn't really wash.
When Greece was trying to sort out its default, it outsourced tax collection to its utility companies as part of the bailout.
Efficient power storage will be a game changer, like a redrawing of international boundaries game changer.
When Greece was trying to sort out its default, it outsourced tax collection to its utility companies as part of the bailout.
Gold is just a commodity, it does not have any magical properties. Since gold is durable, uniform, divisible and portable, it makes a good currency.
Not exactly.
The Greek government wanted to collect a new extra property tax, and needed to collect it as fast as possible (instead of waiting for about a year for the normal tax collection).
So, the payment of the tax was set to be paid temporarily through the electricity bill. The electricity bill is used anyway to collect 1) council tax and 2) compulsory TV licence, and is still used to pay these two.
Now, that extra property tax is paid in the traditional way through the tax office.
The low risk that any breakthrough in batteries would happen IS the reason for the vast amount of securities dependent on it.
Moreover, the cars aren't going away, instead of buying one with a ICE, its going to have a battery and some electric motors, the companies producing that will likely need bonds too. So, without attempting to compute the costs of a few dozen gigafactories, rare earth mines, etc, the article seems pretty worthless to me.
Frankly, i've been wondering for the past few years what percentage of a modern car's production cost is actually the gas engine. Modern engine complexity seems to have peaked 20 years ago, and declined. Of course safety, comfort and entertainment systems have exploded in complexity in that same time period, but those are things people expect out of modern electric cars too.
Drivers of electric cars can choose their electricity source. Whether they want to pay more for green electricity is up to them. If there is demand more green electricity will be produced. With a ICE there is no choice.
>Frankly, i've been wondering for the past few years what percentage of a modern car's production cost is actually the gas engine.
I don't care about that. I've already paid 30% of my cars value just for the gasoline alone. Even if I had to pay 30% more for an electric car then it would still be worth it because of tax benefits and less maintenance.