Disclaimer. After nearly 40 years managing money for some of the largest life offices and investment managers in the world, I think I have something to offer. These days I'm retired, and I can't by law give you advice. While I do make mistakes, I try hard to do my analysis thoroughly, and to make sure my data are correct (old habits die hard!) Also, don't ask me why I called it "Volewica". It's too late, now.

BTW, clicking on most charts will produce the original-sized, i.e., bigger version.

Sunday, November 13, 2016

Tesla Powerwall cost plunge

When Tesla first introduced the Powerwall just over a year ago, it cost US$3000, which was extremely cheap by comparison with all the previous battery options.  Just to remind you, the Powerwall was Tesla's home battery, to store power from your solar panels for use at night, and it had 7 kW of storage, about enough to run an average home in the evening for 2 or 3 hours.  But that price didn't include the inverter (alias the transformer).  Costs for the inverter varied but they were roughly US$2000, so the total cost was US$5000 before installation.  Tesla has just introduced a new Powerwall, which will cost $5500--but, it includes the inverter which is integral to the device, and it has twice the storage (14 kW.)

In effect, in just one year, Tesla has almost halved the price of home battery storage.  In this post, I mentioned how battery costs have dropped 70% in 18 months, as much as they had dropped in the previous 5 years.  Now we are seeing this reflected in retail house-powering batteries.  That's not the only advantage of these new Powerwalls--they're also easier to install; almost plug-and-play, so the installation costs are lower.

So, ignoring the beneficial impact on the grid and the environment, is it worth installing?

According to the NREL PVWatts calculator, 5 kW of solar panels would produce an annual output of 6909 kWh where I live (in Brisbane, 2000 km further north, output would be 10% more, and with much less seasonal variation, but in London, latitude 50 N, it would be just 4,500 kW, 35% less). Average annual household usage in Oz  is 5817 kWh.  Some of the output produced is going to be sold back to the grid in summer.  And in winter, electricity generation is insufficient: you'll need to buy power from the grid.

You can now buy 5 kW of PV panels, fully installed with inverter for A$3900.  US$5500 for the Powerwall 2 is about A$8000, assume installation is A$1000.  Total cost of Powerwall plus 5 kW of solar panels would be A$12900.  Using the tariffs of my electricity provider, Diamond Energy (8 cents feed in tariff, peak 21 cents usage tariff, 12 cents off-peak) and assuming that the battery allows me to use 100% of my solar power in summer, and in winter the battery charges overnight using off-peak power, the annual savings would be $1300.  Ignore interest charges but also ignore the rise in electricity prices, and the cost savings will pay for the battery and the panels in 10 years. The panels will last at least 25 years, and although the battery is only guaranteed for 10 years, it will still have plenty of juice at the end of 10 years.  And remember, battery and solar panel costs are still falling. Right now the combination of 14 kW battery plus 5 kW panels will pay for itself over 10 years.  In 5 years, costs will have likely halved, or better.

But if the Powerwall has nearly halved in cost over 1 year, what about the industrial/utility scale Powerpack?  Well we don't know; Tesla hasn't revealed new costings.  I estimated the LCOE of the Powerpack as $230/MWh, which was still somewhat more expensive than peaking gas ($163 to $218) But assuming the same decline with the Powerpack as has happened with the Powerwall (it's almost exactly the same technology) the new LCOE of the Powerpack will be $126, much lower than peaking gas.  Much lower.  Would you build new peaking power gas generators when batteries are so much cheaper?  You'll keep those you have, for now.  But new ones?  Nah.

And what about cars?  The Tesla model 3, due out towards the end of 2017, was projected by Tesla as costing US$35,000, before incentives  The average new car in the US sells for around $34,000.  Most of the cost of an electric car used to be the batteries, so if battery costs have halved EVs are going to be much cheaper.  Did Tesla expect battery costs to halve so quickly?  Did anyone?  Will Tesla pass the gains on in quality improvements or price cuts?  Or will they pocket the profit?  A possible tactic would be to up the range and the quality on the model S and X, but cut the price on the model 3, while keeping its range the same.  Remember, Tesla has the widest and densest range of fast chargers, and even if they won't be free any more, they still mean that range anxiety with the Tesla is vieux jeu.

From 2017 on, Tesla is likely to gain a 1/3rd or better share of the mid-priced car market, just like its 1/3rd share of the luxury car market.  And other car manufacturers are switching too.  Volkswagen is targeting 25% electric share of its sales by 2020, Toyota started a battery division at the beginning of this year and has quietly conceded that hydrogen-cell vehicles aren't going anywhere, General Motors has the Chevy Bolt.  By 2020, 1/3rd of all new car sales could be electric.   By 2025, 2/3rds, because even small cheap cars will be EVs: they'll have a small range--100 miles/160 km--but most commuting is less than 100 k per day.  These small cars will have small battery packs, with perhaps the option to upgrade.  And batteries by then will be a quarter of their cost now.  And this will mean that even the cheap car market goes electric.  On my calculations. this will mean that demand for oil in the US will fall by half by 2030.  In China, EVs are now 25% of new sales.  Some European countries are considering banning petrol cars by 2025.  Global oil demand and global CO2 emissions will plunge.

This is a tipping point.  Between latitudes 50 N and S, every house, school, factory, shopping centre, or office complex which has enough roof space will have solar panels and behind the meter battery storage. In electricity generation the switch towards renewables will continue.  Wind and solar are now half the price of coal, and as cheap as or cheaper than gas. Gas was uniquely useful to fill the gaps left by variable renewable supply, and to provide peaking power.  But batteries are now cheaper--and better.   And their costs keep falling.  Within 20 years, 99% of the car fleet will be electric. The Right have done nothing to support this essential shift towards renewables.  In fact they have done as much as possible to stop it.  But Trump and his troglodytes are powerless to stop this.  Technology and economics have taken the lead.    From 2018 onwards, CO2 emissions will start to fall, and that decline will inexorably accelerate.

Thank you, Elon Musk.


  1. Even without the assumptions ("drop by half") it's looking good. Don't forget the when a vehicle battery pack is too worn out for the fast and heavy charging required of a vehicle it's still suitable for renewable storage. This will definitely help the economics of electric vehicles.

  2. I don't think any countries in Europe are planning on banning fossil fuels cars in the near future. I think that they're talking about banning sales of new fossil fuel cars.