A few days ago, a good friend of mine pointed me toward a presentation on disruptive technologies given by Tony Seba. A youtube video is available here:
The entire video is worth watching, but today I will restrict myself to the issues he raises relating to battery technology.
Seba stresses that technological change in the transport sector could happen at breakneck speed. With a pair of compelling photos of early-last-century New York, we are asked to remember that a grand disruption in transport has happened before. In the first photo, dating from April 1900, we play a game of spot the car (click for larger image).
In the second, a mere 13 years later, the challenge is to spot the horse.
The lesson here is that once a disruptive technology reaches a particular tipping point, it doesn’t just take market share from the incumbent industry but rather completely replaces it. For Seba, we are close to reaching that point with electric vehicles.
According to his analysis, EVs have three unique advantages over internal combustion engine (ICE) vehicles: rock-bottom fuel costs, minimal maintenance costs and instant power. Consequently, once EVs reach the price points of conventional cars, every sane consumer will opt for an EV.
The critical cost for the EV is the battery. Once the battery price gets below a threshold level, the ICE car is as obsolete as the horse. Does this analysis hold up? First, let’s look at the main current EV offerings in the UK market:
In the UK, the retail price of electricity for residential use is around 12 pence per kilowatt-hour (kWh). In US dollars, this is about 18 cents per kWh. So to fill up a BMW i3 will cost £2.26 (US$3.38). If we take the middle driving range of the different i3 versions to be 80 miles, then cost per mile is 2.8 pence (4.3 cents).
To put this in context, a Mini from the same BMW stable usually has a 40 litre tank, which would cost around £44 to fill at current prices. A full tank for the Mini will allow you to drive around 650 miles, giving a price per mile of 6.8 pence (10.2 cents). So i3’s fuel running cost is 40% of that of the Mini.
Meanwhile, the zero to 60 mph times for the two types of car are similar, which brings us to pricing. The i3 sells for £30,000 (US$45,000) before the UK government EV grant of £5,000. By comparison, a Mini sells for around £15,000 depending on the specification. Against this background, we have to ask whether the i3 battery price will fall sufficiently to get an i3 close to the price of a Mini.
The battery is a composite of the actual cell plus the packaging. Seba’s presentation is a little vague over what he means by the battery price, so I have taken some numbers out of a presentation by Winfried Hoffmann, president of the European Photovoltaic Association.
As a ranging shot, let’s take the current cost to be $400 per kWh. Accordingly, the current BMW i3 battery will be priced at £5,000 ($7,500). Assuming the BMW’s electric motor is around the same price as the Mini’s internal combustion engine, the missing £10,000 is the cost of putting the EV together (integrated electronics, carbon fibre body and so on).
Moreover, to overcome range anxiety and so make the i3 truly mass market, BMW will likely need a 50kWh battery to give the car over 200 miles on one charge (viewed by many in the industry as the magic number). Ignoring battery size issues, I shall add another £7,500 to the current cost of the i3 to give an estimated price of £37,000 for a car lacking in anxiety.
In a similar manner, let’s say that £10,000 is old school technology (I’m assuming the internal combustion engine in the Mini accounts for a third of its price). Therefore, £27,000 is electronic kit, and could, theoretically, be subject to some kind of Moore’s Law dynamic. Of course, the £10,000 of traditional car stuff in common between the two types of vehicle will also grow in sophistication (and likely become more electronics dominated), but this has no bearing on parity calculations.
Now let’s play exponential curves. Seba states that battery price falls have already accelerated from 14% per annum to 16% per annum over the last few years. He believes that this trend is set to continue given that the electronics, automotive and energy industries area all throwing billions of dollars at battery technology. For argument’s sake, I will keep to a decline rate of 16%, rather than have this accelerate, and I am going to apply this decline rate to the electronic kit and new material in the EV.
At these decline rates, I would assume that the EV will reach parity with a conventional car by 2025–quite a lot later than when Seba believes this will happen. But then again, Seba sees cost decline rates accelerating.
As a mind game, let’s assume Seba’s forecast is right. Consequently, within a decade or two we will see a substantial portion of the UK’s 30 million passenger vehicles replaced by EVs. If half of passenger vehicles made the technological transition, then that would give the UK a total of 750 gigawatt-hours (GWh) of storage (assuming an average of 50 kWh of batteries per car). In 2013, the UK generated 357 terawatt-hours of electricity (source: here), or around 975 GWh per day. So the UK’s EV fleet could store around 75% of current daily electricity generation. Should this state of affairs come about, the implications for the UK electricity grid are enormous. And, of course, this phenomenon will be taking place worldwide.
I am somewhat more sceptical than Seba over the timing of any transition to EVs, Yet once batteries–and thus EVs– reach critical price points he is right that this will be transformational.