A recent EV growth report by Deloitte gets key facts wrong, ignores Tesla, and projects EV growth rates to be drastically slowing, glacial, after reaching just 20% of market share in 2030. And these guys call themselves professionals? Let’s help them out.
Widely covered in the tech media, usually without analysis or critique, Deloitte’s recent EV market growth report is full of holes and basic factual errors, which they leverage to support a downbeat assessment of EVs’ long-term prospects.
Since Deloitte cited CleanTechnica as one of its sources (bottom of page 10 of the full report), I felt responsible to help them out of their confusion and set them on a straighter path. I’ve contacted the report’s lead editor as well as one of the key authors, but haven’t yet received any solid explanations for the discrepancies.
The Deloitte report sensibly takes the position that EV mass-market demand will depend on EVs offering acceptable performance in the key areas of: driving range, total-cost-of-ownership (TCO), and DC fast charging speeds (and infrastructure):
“…the four most important customer concerns regarding BEVs are: driving range, cost premium, lack of infrastructure and time required to charge.”
So far, so obvious. Their key error, however, is to significantly misunderstand and misreport both the performance on these key metrics of EVs on the road today, especially the performance of Tesla’s vehicles, and the rate of improvement in these key metrics of EVs coming later in 2019 and subsequently. The report’s errors, confusions, and omissions include:
- Rookie errors and confusions about the range of EVs (and excluding Tesla’s Model 3)
- 5 years behind the curve on the evolution of DC charging speeds
- Errors understanding the critical role of battery cost ($/kWh) on EV range (they instead claim technological hurdles)
- Arbitrary and unsupported analysis of the TCO of EVs, downplaying TCO parity with gasoline/diesel vehicles
Whether these are genuine rookie mistakes or deliberate misstatement of facts in the service of an overall picture of EV FUD is hard to say. It’s possible that they simply misunderstand the current state of EVs and their rapid evolution. It’s also possible that the multi-author, multi-editor nature of the report just failed to catch errors and unrealistic projections. We’ll analyze the errors themselves in more depth below. For now, let’s look at what unrealistic conclusions they lead Deloitte to draw.
In terms of the big picture of EV growth rates, Deloitte gets at least one thing right — the historical figures on the growth in EV sales over recent years. Their historical growth figures match the industry-leading sales reports by our friends over at EV Volumes. Although EV Volumes doesn’t yet have an official report on 2018 total EV sales (at time of writing), Jose Pontes has indicated that it will very likely be over 2 million vehicles worldwide. Deloitte is informed well enough to use this as their 2018 figure. (The EV sales figures include both pure EVs and plug-in hybrid vehicles.)
The chart shows an average 5-year growth rate of ~56.66% year over year. In the period from 2016–2018, the growth rate trends further upwards to an average of 62%. Deloitte knows this, since it has the same data and used it for the report’s charts. What then do they project the future sales numbers will be in the coming few years? Let’s take a look:
Notice that Deloitte’s data for 2017 and 2018 are historically accurate. Any differences here are not due to different interpretation of sales numbers prior to 2019.
On first appearance, this looks like a fairly normal graph. On closer inspection, in plotting the data as the company does, Deloitte is assuming that the historical growth over the past several years — up to and including 2018 — is about to drop off very significantly.
The green data show the historical growth rate (56.66%) continuing over the coming years. For Deloitte (the red data), it’s as if the growth up to now was a fluke and, after 2018, Deloitte projects that growth slows and slows. You can see that, for 2019, Deloitte’s red bar is immediately down on the recent growth trend projection depicted in the green bar. In figures, its growth projection drops from the recent 5-year average of 56.7% to just 37.5% in 2019. From there, it only gets worse in Deloitte’s prediction. Here’s the company’s complete 12-year forward projection in year-on-year % growth terms, plotted against the year-on-year % growth rates that have historically been recorded:
Deloitte sees EV growth rate immediately dropping from recent 5-year 56.7% year-on-year average and quickly trending to zero! Down, down, down! Doom and gloom. There’s a small uptick of growth around 2023 and 2024. That’s around the time they claim EVs will achieve TCO parity with gas/diesel vehicles (we at CleanTechnica think it’s much earlier, discussed below). Deloitte predicts around 9–11% market share for EVs in 2023–24 (below graph). After this, however, EVs apparently must get relatively less attractive, because EV growth drops off a cliff, trending into single digit % growth after 2027, according to Deloitte. This declines further, trending towards zero growth from 2030 onwards, after EVs’ overall market share has approached 20% (see graph below).
Why does Deloitte make these predictions? Is there a reason why its analysts think EV growth grinds almost to a halt after 2030? Perhaps they think reducing carbon emissions will no longer be a priority, we’ll all come to accept inevitable unbridled climate change by that time, and want to accelerate ourselves off the edge of the abyss? Or perhaps they think we’ll just miss the noise, vibration, slow response, mechanical complexity, and high running costs of internal combustion engine (ICE) cars and turn back towards them?
Here’s a graph of Deloitte’s complete EV sales projections out to 2030, compared to a projection based on recent growth trends and including a conservative middle-ground position that I’ve pencilled in:
The grey bars in the chart represent Deloitte’s projection of EV manufacturing capacity over the coming years. It’s clear that they think demand for EVs, and thus sales (the red bars), will be substantially lower than the manufacturing capacity. You can see clearly their assumption that any growth in EV sales will taper off towards 2030. On the basis of these projections, Deloitte’s final conclusion is therefore to warn automakers that there won’t be sufficient demand for EVs going forwards, and thus their investments for EV manufacturing capacity are unwise. Perhaps Deloitte’s report is aimed at ruffling some feathers and getting automakers to turn to them more frequently to get their cautionary advice on the future of the auto industry.
No one knows for sure what the future holds. We try to make reasonable assumptions based on what we know of the past and the present, and the trends and dynamics that we see emerging. We can, however, take a look under the hood of Deloitte’s assumptions and understanding of the key dynamics, because the firm shares its assessment of these in the report. I agree with Deloitte that capable range and recharging, at competitive cost, will be key to EVs gaining ever more market share. As I mentioned above, however, there are many glaring errors in how Deloitte analysts read the current status of these aspects, and glaring errors in their awareness of the current evolution of EVs. Let’s break down their errors in more detail.
Rookie Errors & Confusions about the Range of EVs (and Excluding the Tesla Model 3)
Deloitte analysts are noticeably inconsistent in reporting the range of current and upcoming EVs (page 7 of their report). They state that they are using (the outdated) NEDC range, but their data set actually uses a mixed array of metrics, and this mistake made it through the editing process. It’s a hotchpotch of the more realistic EPA range metric, the unrealistic WLTP metric, and the essentially useless NEDC metric. For example, they list the forthcoming Tesla Roadster as having 1,000 km (620 miles) of NEDC range, when in fact this will be its EPA range, a very different metric.
Whatever metric Deloitte uses or confuses, it reports the Audi E-Tron (512 km) and Mercedes-Benz EQC (500 km) as having greater range than the Hyundai Kona EV (420 km) and Kia e-Niro (470 km). This doesn’t match these vehicles’ official NEDC ratings nor the WLTP ratings (e.g. Kona 449 km WLTP, 546 km NEDC), and most likely also misrepresents their relative performance based on their upcoming EPA ratings. For example, the Kona has an EPA range of 258 miles, the Niro 239 miles — the reverse of Deloitte’s ranking. It’s highly doubtful the e-tron or EQC will be EPA rated at as much as 258 miles EPA.
Deloitte also fails outright to list the Tesla Model 3 or Tesla Model S, which are widely known to be the current range leaders of all EVs. No reason is given for this. One reason might be that Deloitte’s clients are mostly European automakers who have already had enough of hearing about how sexy Teslas are.
5 Years Behind the Curve on DC Charging Speeds
Deloitte state on page 9:
“We estimate the time required to achieve an 80 percent charge in a 60 kWh battery (equivalent to a range of 200 miles) will be reduced to just 30 minutes — the threshold the majority of customers consider acceptable — by 2025.”
In reality, if we’re setting the bar at recharging 160 miles (80% of 200 miles) of range using the generous NEDC metric, the 2019 Hyundai Ioniq, coming in the second half of this year, already fits the bill, more than 5 years ahead of Deloitte’s projections.
Its NEDC range is 235 miles (378 km), and it can recharge 80% of its 38.3 kWh battery pack in under 30 minutes. Its more realistic EPA range rating will likely be around 170 miles, however.
Then there’s the Tesla Model 3 Long Range and Mid Range. The Long Range can already add 198 miles of range (EPA) in 30 minutes, and more than 200 miles in Deloitte’s preferred NEDC-equivalent metric. The Mid Range can add 177 miles (EPA) in that time. Perhaps Deloitte mean that the 80% charge is the equivalent of 200 miles range? When the Tesla Supercharger v3 arrives later this year, both these Tesla models will likely achieve over 200 miles added in 30 minutes.
And then there’s the Volkswagen ID, arriving in late 2019 or early 2020, all variants of which are claimed to have over 200 miles of WLTP range (let alone NEDC range) and capable of adding 80% charge in under 30 minutes.
In short, the Deloitte report’s “estimate” about 2025 charging speeds is already out of date based on what the Tesla Model 3 could do in 2018, and what several other EVs will do in 2019. Adding 160 miles of range in 30 minutes is already becoming the new normal in 2018–2019 and 200 miles is what today’s leaders can achieve. By 2025, the norm will be adding 200+ miles in 20 minutes (or better). What was Deloitte thinking by listing 2018–2019 specs as an important goal for EV acceptance in 2025? This aspect of the report appears to be totally out of touch with reality.
Missing the Role of Battery Cost on EV Range, and Suggesting Technology Breakthroughs Required
Keen on the idea that 200 miles of range is a useful performance to aim for, the Deloitte report gets creative on page 7, giving us a helpful sounding section titled, “Next generation BEV driving range explained.” It points to evolving battery materials and cell chemistries, optimizing pack designs, pack cooling, and BMS improvements. It even looks beyond 2023 to exotic possibilities like lithium-air batteries and solid-state electrolytes. All sounds very exciting!
Presumably, then, giving EVs decent 200+ mile range depends first and foremost on these scientific and technical improvements and breakthroughs? It’s a question of physics more than one of cost? Well, no.
Tesla has already proven that 300+ mile range is available in their mid-sized Model 3 sedan (in 2017, actually). Smaller still, the Chevrolet Bolt (2016), Hyundai Kona EV, Kia e-Niro, 2019 Nissan LEAF, and upcoming 2019 Volkswagen ID all have ranges comfortably over 200 miles, and over 250 miles in some cases. Obviously, the energy density of batteries is already past the point of allowing very decent range in these modestly sized vehicles. It’s not primarily a question of physics and technical breakthroughs, although chemistry improvements can reduce costs by making the same amount of raw materials “go further” in energy terms. But the overarching point is that long-range EVs are not a challenge in technology terms — they are primarily a challenge in simple cost terms.
Whilst gradual improvements to chemistry can help reduce cost, more relevant are optimizing manufacturing processes and volumes, optimizing cell geometry, etc. These important priorities are nowhere discussed by Deloitte, but luckily it’s a topic that we discuss regularly here on CleanTechnica.
The short version is that battery prices ($/kWh) have been reducing at around 15% per year for the past few years, and we expect this to continue onwards, with Tesla targeting breaking through $100/kWh at the pack level around the end of 2020. LG, CATL, and other battery makers are also likely not too far behind. Does Tesla spend much time talking about battery chemistry breakthroughs? Not at all. It has actually dismissed the idea that breakthroughs are necessary (gradual improvements, yes, but not breakthroughs) — it’s all about increasing scale and steadily reducing costs.
Arbitrary and Unsupported TCO Analysis of EVs, Downplaying Parity with Gasmobiles
Deloitte’s claim that an EV compact car driven 7,900 miles a year reaches TCO parity with a gasoline/diesel car by 2022 (actually, they say 2021–2024, depending on incentives) is not accompanied by any hard data on cost calculation to back it up. Our own analysis here at CleanTechnica (and here and here) has shown that, in many scenarios, EVs’ TCO was already better than gas cars’ TCO in 2018 for mid-sized segments (and pretty much every segment above). This is in the US, where the gas-price to electricity-price gap (a big part of EV TCO savings) is relatively low compared to other regions. Without Deloitte’s TCO claim being supported by clear data, we can only judge this aspect of its timeline to be downbeat compared to other analyses.
The largest part of TCO is a vehicle’s value depreciation. ICE cars are on the cusp of having a miserable future in this respect, just as the current generation of all-round competent EVs are topping the value-retention charts. Value retention is only going to get better for EVs and worse for ICEs from here on out. This is also a topic we’ve regularly covered. More broadly, given the inexorable battery cost reductions mentioned above, we can expect pretty much all mainstream vehicle segments to be at sticker price parity with ICE by 2024. Given the significantly lower running costs of EVs, this puts TCO parity far in advance of 2024 for most vehicles.
Deloitte prides itself on being the largest “professional services network” in the world by revenue, and its recent report presents itself as a forward-looking and authoritative view over the future of the EV market. In my view, however, based on what I’ve presented in this article, the quality of the data in the report is poor and contains significant errors. It appears to be badly researched, and full of perspectives that resemble rookie mistakes. Based on the erroneous data, Deloitte has anaemic projections about EV growth. Its conclusions are downbeat, yet unsupported, unjustified, and out of step with realities on the ground. Given that the conclusions could only be justified on the back of the report’s erroneous data, this report certainly has the result (intended or not) of promoting misguided fear, uncertainty, and doubt (FUD) over the future of EV sales growth.
Similar reports promoting FUD about EVs have been published before, and will likely increasingly appear over the coming years as EVs inevitably capture more and more of the market, and disrupt more and more of the legacy auto industry. It’s best to either ignore them or read them with a critical mind.
If you can spot more errors in the Deloitte report, please share your findings (and other thoughts) in the comments.
All images via automakers unless otherwise stated.