How and When Will Electric Cars Replace Fossil-Fuelled Cars? Part 1

Figure 1: Diffusion of Innovations Curves (Image: Wikipedia)
Figure 1: Diffusion of Innovations Curves (Image: Wikipedia)

Electric cars are here, now, and selling in increasing numbers. They are getting better and cheaper each year to the extent that they are typically only slightly more expensive to buy than a fossil-fuelled car, and yet they are much cheaper to run. EVs are quieter, smoother and have zero emissions, and increasingly have the performance and range to match fossil-fuelled car.

It is natural, therefore, for the question to arise as to whether, or when, electric cars will replace fossil-fuelled cars. A quick Google search will show up many opinions on this subject. EV drivers are often very optimistic and believe EVs will take over in just a few years (say, by 2020). In contrast, conservative organisations don’t just take an opposing view, they often seem to ignore recent developments and seem to be able to argue that EVs will never compete with conventional cars or sell in significant numbers, even though they are already doing so.

An interesting example of this doublethink is the U.S. Energy Information Administration’s ‘Annual Energy Outlook’ report for 2014. It is due to be released soon, but an early version of the report was released at the end of 2013. It includes the headline prediction that in the US Just One Car Out of 100 Will Be Electric in 2040. This despite the fact that US plug-in sales have already increased from 0.14% in 2011 and 0.37% in 2012 to 0.62% in 2013 (source: Wikipedia).

It seems that most analyses of future EV market share simply represent the opinion of the author. However, I believe it is actually possible to make a data-driven projection of future market share based on known market share data so far. Further, the adoption and growth in market share of innovative technologies (computers, mobile phones, tablets, etc.) is already well studied and I believe EVs will follow the same path.

Specifically I posit that EVs will follow the standard diffusion of innovations approach, which is a theory that explains how, why, and at what rate new ideas and technology spread through cultures. A key element is that adoption begins with a small subset of buyers, the Innovators, then passes through other categories of consume to achieve full adoption (the other categories are Early Adopters, Early Majority, Late Majority, and Laggards).

The rate of change of market share follows a ‘bell curve’ with adoption being very slow at the start, then increasing significantly, then levelling off at 50% market share, before falling away as it takes over the remaining market – this is the blue curve shown in Figure 1. It is interesting to note that within the rate of adoption there is a point at which an innovation reaches critical mass – this is a point in time when the number of individual adopters ensures that continued adoption of the innovation is self-sustaining.

If we consider the percentage of market share over time then this follows an S-curve, with very low market share expected at the start. This starts to accelerate as the majority groups adopt the innovation, then it slows down again as the market approaches saturation – this is the yellow curve shown in Figure 1.

The adoption of electric cars is currently still in the almost flat starting part of this curve. In the next part I will show how we can apply the data we have on EV market share so far to this adoption curve to predict future adoption rates and market share for EVs, and calculate the likely timescale until EVs dominate the car market.

[Part 2]

Home Forums How and When Will Electric Cars Replace Fossil-Fuelled Cars? Part 1

This topic contains 12 replies, has 3 voices, and was last updated by  donald 6 years, 2 months ago.

Viewing 10 posts - 1 through 10 (of 13 total)
  • Author
  • #8164

    Trevor Larkum

    Electric cars are here, now, and selling in increasing numbers. They are getting better and cheaper each year to the extent that they are typically on
    [See the full post at: How and When Will Electric Cars Replace Fossil-Fuelled Cars? Part 1]



    Well, it’s not that hard to predict the future. Look at Norway, a country where EVs are cost competitive with ICE cars. In March 10.8% of ALL cars where Telsa Model S cars. And about 3% where Nissan Leafs. That’s about 14% with just two EV models. That is proof that EVs will take over as soon as they are prices similar to ICE cars. And that day is there in Norway, and drawing closer in all other parts of the world. Not question in my mind is not ‘if’ but ‘when’


    Trevor Larkum

    I’m guessing you enjoyed the official prediction of 1% market share in 2040!



    “Electric cars are here, now, and selling in increasing numbers. They are getting better and cheaper each year to the extent that they are typically on
    [See the full post at: How and When Will Electric Cars Replace Fossil-Fuelled Cars? Part 1]”

    I can see the logic, but the devil is in the detail. Firstly, I don’t understand the numbers noted on the x-scale you are working to – the rate of change of market segment would be in units of, e.g., %/year.

    Secondly, the inference is that the new technology *can* replace the existing. Usually for this sort of technology penetration, the new product improves on the performance of the old. That is not true of EVs and I do not think it is yet known when we will know if it can do that at all.

    The problem is petrol – if you were to design the most energy dense means of carrying energy that is easy to liberate from the energy vector, you would not end up with a battery! You would design a hydrogen-rich fuel that is liquid at STP. You have this as a moving goal-post after that because as you increase the attractiveness of EVs then you incentivise the selling of hydrocarbon fuels at lower and lower prices because if it isn’t wanted then it will be sold for less to keep up a customer base.

    Ultimately, if electricity and petrol competed head-to-head without Government biases, then you have pre tax/duty costs of 58p/litre (approx 6p/kWh) for petrol and double that for electricity at ~12p/kWh. Now the EV can make better use of that energy in summer, but ultimately if it wasn’t for the tax/duty then the per mile costs are comparable.

    Anyhow, my point is that an EV cannot yet replace an ICE because it just can’t compete on range and refuelling rate, so the concept that the whole national fleet of ICEs will be replaced with EVs is not yet a reality that could happen.

    I think what we will see is a pause, as you describe, followed by a fairly rapid uptake of EVs by multi-car owning households with off-road driveways. In actual fact, this is a significantly large fraction – I have read around 50% of households have at least two cars.

    So I imagine your blue curve will show some interesting characteristics around 25% or so (that is, the percentage of cars in the 50% of multi-car households where one is replaced with an EV) and then the take up of EVs will stagnate for a while before then increasing at a more glacial pace to a total take up yet to be determined by how far it is really possible to get a maximum take-up of EVs.

    The whole of the prediction curve beyond 20% take-up is, I would say, predicated on the introduction of technology that is not yet available, let alone ready. That doesn’t make it impossible to predict if larger take-up will happen, but it does make it impossible to say ‘when’.



    Note, where I said “I think what we will see is a pause, as you describe” I was implying not a pause in the takeup, but in the rate of change of take up (i.e. a slow uptake as we are seeing now).

    (I can’t edit the post, there seems to be something wrong with the website, it is giving me some configuration error when I try to edit.)



    I don’t think it’s fair to say EVs can’t compete with ICEs and never will. If you can refuel a Tesla Model S in 30 minutes for 300 miles (for free, I might add) today, how is that not copeting. Taking a 30 minute break after each 300 miles is simply not a problem. You do need to strech your legs, go to the toilet or take a bite anyway. The only problem is cost and range. Range has been solved by the Model S. Cost has not yet, but if Tesla comes through with their promise of cheaper EVs in just a couple of years, I expect EVs to be cometitive with ICE cars on both price and range just a couple of years after that. They already are very competitive on cost per mile and maintenance.

    And hydrogen, really? Have you seriously looked into why we don’t have hydrogen cars yet? There are good reasons for that.



    No. I didn’t mean ‘hydrogen’, just hydrogen rich fuels; i.e. petrol and diesel. I’m saying that if you had no previous idea on how to power a car and decided to invent something to put as much energy as you could into an energy vector, you would invent a hydrocarbon fuel. Just so happens we’ve already been lead to that point.

    To clarify, I would not say that EVs will never compete with ICE, but they don’t right now. There are two reasons why not;
    1) So you think £68,000 competes with a £35,000 Jaguar? (I name Jaguar simply because you will probably agree the Tesla is a similar class. I have been in a Tesla and, actually, I don’t think it is as ‘premium’ as Jaguar is)? The difference in purchase price between a Tesla and an equivalent ICE would pay for all your fuel for a typical lifetime of that car, and also leave you a fat deposit for when EVs really can compete with ICE. OK, the base model with no options and 200mi summer range comes in at £50k, but the supercharger option alone costs the same as fuel for the XF 3.0D for 12,000 miles, and the £15k small change would still cover 100,000 miles worth of fuel. I’m not saying it is way off, but it isn’t competitive yet because the EV is reliant on a weak and embryonic infrastructure.
    2) Practicability:- If ICE were replaced with EVs, that’d be 30 million cars needing recharging, on average likely once per day. There are (a staggering) 140,000 vehicles per hour on parts of our motorway network. Assuming they will stop for a recharge every two hours, at 70mph, you’d need approx one charging point every 140 miles to service the cars running on the motorway. So you need, on average, 1,000 chargers in each mile of motorway….. How many ‘superchargers’ are Tesla planning on?


    Trevor Larkum

    Donald – thanks for the feedback on the post, but note I’m just explaining the theory for now and I’ll post charts with actual EV data in later posts.

    With regard to your comment on 140,000 vehicles per hour on the motorway, I think it’s missing the point. If everyone has an EV with a range of 300 miles and they charge it overnight then the question is how many would ever need to charge on the motorway. Based on my experience of how often I drive more than 300 miles in a day I would say it’s a tiny fraction (perhaps 1%, perhaps much less). So of the 140,000 cars that drive past a particular services maybe 1400 need to charge once during their 600 mile trip, implying you might need 2-3 chargers (not 1000) per mile of motorway (or perhaps 20 at each services). The real figure is probably somewhere in between.



    I agree that not all cars will be on routes requiring charging, but nor would I imagine it is as low as 1%.

    There are, clearly, a number of unknowns in this, but some judgements are needed to get a handle on the scales.

    Some motorway lengths I agree are not going to be carrying ‘long distance’ traffic in large concentrations, which will be ‘urban’ type lengths, e.g. the extended Birmingham or Manchester conurbations and environs.

    However, some will take large volumes of long distance traffic, M4 and M5 to the south east, M40 south of Oxford and M6 north of Preston spring to mind. I doubt much of that traffic is on anything other than a long distance drive.

    Looking at the debacle of the recent changes in Chargemaster and PIM, and the poor reliability of the rapid charger network, one can only wonder if these are teething troubles or a sign of things to come. I think it will be a mixture of both.

    One way or another, it will be a competitive matter. Money will be charged for charging, and if there are too many takers then the price goes up. This is normal supply and demand. Longer distance EV drivers will be captive on the motorway in a way that longer-range capable ICE drivers are not, because there will be an increasing rate of diminishing returns heading off the motorway looking for cheaper charging, which itself consumes the precious range one is hunting for. One can already imagine that there may be dedicated ‘charging farms’ close off key junctions of the major routes that accommodate 10,000 EVs, because there just isn’t enough space in the current set of motorway service areas. All these extra costs and, more important to some to much time taken over the route, will balance the benefits of ICE with EV.

    For these reasons, and my above opinions, I remain fairly certain that EVs could ramp up very quickly in the next 5 to 10 years or so, but they will hit a dwell point of 25% take-up for a decade or two as the realities of mass EV ownership begin to become understood and the problems managed and dealt with.



    Another aspect is where the energy comes from. Total consumption of fuels for cars and small vans is around the 30 million tonne mark per year.

    Converting that into MWh based on 42MJ/kg, we get 3.6×10^8 MWh – 360 TWh used.

    This is equivalent to the current entire electricity consumption of the UK.

    Now, let’s estimate EVs need 1/3rd of that actual calorific energy because they are more efficient (I think it is more like 1/2 because of inefficiencies in electricity transmission, but let’s go with 1/3rd so as not to be accused of making it look worse). So we’ll need 120 TWh per year to drive the current UK fleet if they were electrified.

    In 2012 we, in UK, generated 70 TWh of nuclear power, 19.4 TWh from wind, 8 TWh hydro and 17 TWh from biofuels and wastes, with 100 TWh from gas & 144 TWh from coal.

    So to achieve an additional 120 TWh from non-fossil fuel based we have to double ALL of the current nuclear and renewables generation capacity. Or to consider that in nucelar-only terms, an extra 15 GW. Say, an extra 15 Sizewell B’s needed, or about one nuclear power station per 2 million cars.

    Or to generate that energy from wind turbines, consider the 175 turbine ‘London-array’ off the Kent coast…. a peak rated power of 650 MW which (by my estimate) should give around 1 TWh of annual energy. We’d need 120 lots of the London array to feed an EV fleet replacing ICE, an extra 21,000 turbines, about 1,400 cars per turbine.

    (Arithmetic error is always possible in such estimations. Please feel free to point out any errors I have made!)

    Nuclear seems the only realistic way. EVs are the best advertisement for nuclear power there has been!!

Viewing 10 posts - 1 through 10 (of 13 total)

You must be logged in to reply to this topic.