Electric Vehicles: Fruitless Search For An Impossible Dream?

Includes: DUK, ED, F, GM, PCG, SCG, SO, TSLA
by: Gary Golnik

Car companies may not see profits from EVs.

Electric utilities are not structured for an EV future.

Operating costs of EVs may be much higher than supposed.

Automotive and utility stocks are not the best place to put your money

Most auto companies and utilities are investing substantial capital in the development of electric vehicles and the electric grid structure to support them. These investments are certainly trendy, but they are not financially sound for the long haul, because they rely on government and societal investment that is extremely unlikely. Both auto companies and utilities will be difficult investments over the next decade, with a clear path unlikely to emerge. This will translate into sluggish growth and reduced profitability.

The cost constraints for EVs suggest that car companies can have a clean vehicle, or a profitable one, but not both. The movement to EVs is mainly psychological, which leaves car companies vulnerable to a shift in consumer opinion. Utilities are in a somewhat different place, but face the daunting prospect of raising huge amounts of capital to develop the infrastructure required to support EVs. This investment will be in two primary areas: developing a "smart grid" to handle a very different proposed generating structure, and the huge increase in the development of generating facilities required to support EVs.

The system costs of EVs are being ignored

In order to realistically evaluate an electric future, we need to look at the entire system, not just one small part. Many years ago, as an engineer, I evaluated the potential for very large electrical lasers. They looked really good (and efficient) if I only looked at "wall plug efficiency". That means everything downstream of the wall plug. Unfortunately, what comes out of the wall isn't the whole story. The same is true for electric vehicles. Let's look at some of the real costs to provide the electric dream, convert them into an equivalent gasoline cost, and compare them to traditional vehicles.

Some ground rules for the comparisons: Let's assume that the average car is driven 15,000 miles per year (the US DOT shows 13,476 as of March 2018 averages, Average Annual Miles per Driver by Age Group). More miles will make EVs look a little better, so 15,000 is slightly biased towards EVs. Let's also assume a 10-year lifetime. This is reasonable because the current average car lifetime is 11.6 years and most battery warranties cover ten years. A further assumption is that, outside of initial cost, the maintenance, repair, tax costs, etc. are the same for EVs and hybrid/conventional. Since we are talking about wholesale replacement of the entire US fleet, government incentives are excluded. I should note in passing that current 5-year resale value of EVs is abysmally low, and this analysis assumes that the issue goes away as EVs become more broadly accepted and battery life data builds up.

Comparison of costs, EVs and conventional vehicles

For conventional cars, current technology is about 35 mpg (10 current vehicles do better). Hybrids are currently up to 53 mpg. I will use 35 and 50 for comparison. As a reference, the current average US fuel economy is approaching 25 mpg. The table below shows the fuel cost per mile for gasoline-powered vehicles.

Cost per mile for conventional vehicles

15,000 miles per year

35 mpg

35 mpg

50 mpg

50 mpg

Gasoline price/$ gal

Total cost

Cents per mile

Total cost

Cents per mile





















For electric vehicles, the future projections are much less clear, but I will use some estimates that I think are reasonable. Three things must be included in the estimate: the electricity cost per mile driven, the initial cost premium for EVs, and the cost of electric grid improvements.

The current Tesla projection is 85 KHW for a 265-mile range. This works out to 0.32 KWH per mile. Electric rates have a big impact, so I will tabulate some variations below. EVs are somewhat more expensive to purchase. The best study that I can find suggests that the EV price premium will likely settle at about $5,000 per vehicle (Breetz and Salon, ASU, "Do electric vehicles need subsidies? Energy Policy, 120, 238-249" Breetz and Salon article). Over ten years, at 150,000 miles total, that's $0.033 per mile. By the way, for those who trade in their cars every 3 years, the cost would go up to $0.11 per mile.

The current electric grid cannot begin to support an all-electric transport network. Estimates vary widely, but the current grid replacement cost is somewhere in the vicinity of $5T (The old, dirty, creaky US electric grid would cost $5 trillion to replace. Where should infrastructure spending go?). Transportation uses 29% of total US energy consumption, while electric generation uses 38% (data from EIA). To replace the petroleum-based transport, we would need 29/38 or 76% more electric generation. The cost of a revised grid is thus 0.76x5T or about $3.8T. I'll further assume that the $5T cost includes the cost of money so that we can divide the 3.8T by 40 years amortization, getting $0.095 T per year. US vehicles drive about 3.22 T miles per year. Expressed as dollars per mile, this is .03/mile.

I think that this assumption is optimistic for EVs. The current grid is minimally renewable (11 % is renewable, but of that solar and wind are only 3%). Building out a new grid (generation, transmission, conditioning) of a more advanced nature is likely to be more expensive than replacing the current grid. Nuclear is off the table due to psychology, oil/gas due to climate, and hydroelectric due to lack of available sites and environmental concerns. We thus need solar/wind that is 33 times larger than currently installed or 58 times larger for the all-electric transport case. Given the intermittent nature of both, the grid would logically be more expensive, although it will probably be even more expensive due to the relatively low thermodynamic efficiency of effectively smaller power stations. The balancing cost alone may be a large driver because of the intermittent nature of both solar and wind.

Electric utilities are not well-positioned to build out a new and advanced grid. How do they raise trillions of dollars? The total combined market cap of the top 10 utilities is very roughly $400B. In 2018, the average utility debt to equity ratio was under 1. How could it ever be 9, which would be needed to raise the 3.8T estimated above? Will the government build the grid and then give it away (there are precedents in the grand hydro projects of the 1930's)? Who would operate it? For all these reasons, I think the cost estimate of $0.03 per mile is low.

For EVs in total then, we have the numbers shown in the table below.

Equivalent gas price for EVs

Electric cost per KWH

Purchase premium

Electric grid upgrade

Mileage cost


Equivalent gasoline cost/gallon 50 mpg gasoline cars



















3-year trade-in - .30 KWH






Pessimistic case






EVs are unlikely to be cost-competitive

I've placed both conventional and EVs on a cost spectrum relating to the equivalent cost per gallon of gasoline. The table above shows that, even at low electric rates, EVs are marginally cost-competitive (the equivalent of $4.75 gas). It gets worse at real rates. Electric costs are shown for several rate levels, and it's important to understand real rates, not "supply costs" or some other fraction of the total. My last electric bill here in Massachusetts was 0.252 per KWH (winter rate). This places the effective cost per gallon of an electrified world for me at about $7.15. EV advocates often use pricing which does not reflect real rates. Very few states in the US have electric costs near 0.10 per KWH.

By the way, remember all that other electricity you use besides the EV? Think those rates are going lower when all of the current generating plants are decommissioned? Washington state has some of the cheapest electricity in the country due to the Columbia River hydro plants. These are clearly not scalable, don't represent full consumer cost, and are still nearly 0.10 per KWH. So the table above seems reasonable.

We have all seen what happens when gas costs go above $4 per gallon. Is it likely that the economy can sustain $7 per gallon gas price equivalents? No! Particularly since I think this estimate is conservative (cheaper) than the reality will turn out to be. The last row in the table is arguably more likely. The future will clearly be some hybrid of electric and gasoline vehicles for a long time to come, but these numbers are a good "limit exercise".

I haven't addressed another serious issue: Americans are in love with large vehicles, and all of the EV scenarios postulate a government-forced return to only small cars. Many of the cost differentials scale poorly with vehicle weight (and certainly expense). Cost differentials for pickups and other large vehicles might easily be as high as $20,000.

What about climate change?

Advocates of EVs assume that they are inevitable (because necessary to help with climate change). So where are we on the whole climate thing? Won't we die if we don't all have EVs? Will the grand scheme to save the planet suspend the rules of financial systems and make the auto companies and utilities more profitable?

Let's look at some numbers. A few assumptions: CO2 concentration is a valid surrogate for overall climate change effects, China is a good example of development, and the world population will stabilize soon at relatively current levels. I note in passing that the worst greenhouse gas, responsible for 70% of the total, is dihydrogen oxide (that truly dangerous substance called water). Also that clouds are the main mechanism, and that clouds are the most poorly understood of all the major factors in climate models. Ah, perhaps another article . . .

The world population is about 7.7 B. This can be divided into the following groups (all figures in the population and 2018 CO2 are rough approximations based on Wikipedia articles):

World CO2 emissions


Population (B)

CO2 emission (Billion tons) 2018

Emission post industrialization

US, EU, Canada
















South America




Asia other




Other inc. shipping




World total




I've estimated a possible future in the right-hand column and in the doughnut chart, assuming that relatively poor countries in the world will not be satisfied until their standard of living (and thus CO2 emissions) are comparable to China's. This is another of the UNs dirty little secrets since they can't openly admit that the poor have to stay poor in order to control climate change. We know that China's industrialization from 2000-2019 caused CO2 emissions to rise from 3B tons to 11B (CO2 emission by country), and I assume that China will peak to 12B (at least). Such a pattern of increase might thus be approximately a factor of 4, and the other regions of the undeveloped world were scaled by that factor, except Asia other, which includes some industrialized regions such as Japan and Korea. Other, including world shipping as the biggest single item, was only rounded upward. US, EU and Canada decrease slightly, following current trends without major economic changes. The likely outcome is a near-doubling of the yearly rate of emission from around 32 GT to about 63 GT.

These values are very hard to convert to global temperature increase because of the uncertainty of the various models, but a molecular analysis suggests that about 8GT of CO2 emission is needed to raise the atmospheric CO2 concentration by 1 ppm. The following article provides some background for the estimates (CO2 background).

How bad is the climate problem

At the level of 2018 emissions (36 GT), this would suggest about 4 ppm per year. Measurements at Mauna Loa from 2016-2018 (Mauna Loa data) show an increase of about 2 ppm per year because some of the emitted CO2 does not stay in the atmosphere. Let's use a range of 2 ppm to 8 ppm to bound the problem. Over 50 years at 63 GT emission (as in the table above), we would be at 450-660 ppm CO2. The IPCC suggests that doubling the CO2 concentration to 560 ppm, relative to the pre-industrial baseline of about 280 ppm, would raise temperatures by 2 - 4.5 C. You can draw your own favorite conclusions, but the UN IPCC has said that anything over 1.5 C is going to be a big problem. My big problem, I mean something that people are willing to spend lots of money to fix.

Can EVs play a significant role?

The US, EU, and Canada now contribute about 25% of emissions, which would drop to 15% in the scenario of the table. Estimates for CO2 reduction for EVs vary wildly, but since transportation as a whole uses about 29% of the total US energy, let's consider the limit where the CO2 emission from all transportation drops to zero. This would reduce the 15% emitted to very roughly 11%, for a total effect on the world's CO2 emission of 4%. Simply put, people won't pay for such a large cost to have such a small effect, particularly when they see their electric bills including the effective cost of $7 per gallon gas. The sticker shock is hidden a bit better now because it isn't advertised at every street corner, but why won't we see electric price ads at charging stations?

Going to EVs will not save the world, and indeed, may not even help except at the margins. Reminds me of the old joke: drunk drops his keys and starts looking for them. Cop comes by and tries to help the drunk who is searching under a streetlight. "Where did you drop them?" "Over there, down the street a bit." "Why are you looking here?" "Cause that's where the light is". Focusing on the US and EU as the problem is the equivalent. Sure, we were there first, but we can't tell the rest of the world not to improve their standard of living, and we can no longer be the only solution to the problem. Replacing coal and oil-fired generation with natural gas will also help get the job done; that conversion has been the main driver in the 16% reduction in US CO2 emissions since 1990. Beware the politician and politicized scientists, who express the data in CO2 emissions per capita to show that the US is "THE problem". We are not and we cannot be the only solution.

EVs are not the best solution

So far, I have tried to show that EVs are too expensive and that they don't solve the problem of climate change. Aren't they going to be adopted anyway as the best of a series of bad choices? Perhaps, but there is a simple and cost-effective alternative. It's even been demonstrated experimentally. In technical terms, it's called albedo management (or solar radiation management). In simple terms, it means that if you are too hot, get out of the sun. In 1883, the Krakatoa volcano erupted. This put large quantities of ejecta (mainly sulfur dioxide) into the atmosphere. The result was increased reflectance of the sun's energy, and less light got to the earth's surface. The following summer was perhaps 1.2 C cooler than average. Global temperatures and climate were unstable for a number of years, but Krakatoa was a huge hammer to the climate. Recent scientific studies suggest that we can do the same sort of thing, with more benign chemistry, deployed by airplanes, at a large cost, but a tiny one compared to the trillions of dollars proposed for a green economy. It's also very selective and selectable, both in terms of the regions of application and amounts. It's not a panacea and a lot of research is needed, but as a cost-effective solution it can't be beaten. An early paper estimates a cost of perhaps $8B per year. Recall that the cost of the revised electric grid alone was about $95B per year. Global costs of green initiatives have been estimated at many times higher than that cost.

Imagine then a future where climate control is a small budget item (NASA's budget is $21.5B per year for FY2019, so they would only need to go to about $30B). Would anyone drive an EV at an equivalent gasoline price of $7 per gallon? We could spend our resources on climate problems that may be much worse in the long run, including groundwater depletion and the plastic sludge accumulating in the oceans. Perhaps the worst news is that focusing on the wrong approach cannot solve the problem. EVs are the wrong approach. They are a band aid on a cancer patient.

Supporting EV policies

From an investment point of view, keeping EVs in the game includes government subsidies and climate change scare-mongering. EVs also have the current benefit that low deployment numbers have hidden the potential problems with the grid. TSLA will be an initial grand experiment, but without subsidies, there is no near-term game, and without huge investments in the electric grid, there is no mid-term game. Only government policy forces the other automakers to stay in the game, and that will fade over time as the fiscal reality sets in. Still, the current situation is all about advertising (which propaganda is). Governments and well-meaning scientists have convinced the world that an immediate crisis exists. Since they are bureaucracies, they have naturally tended towards expensive, brute-force approaches, the better to build a larger group of bureaucrats. The primary bureaucratic lever is the mileage mandate, although other measures to force people to EVs have been considered.

How bad is it for auto companies?

The situation is a looming potential disaster for car companies, albeit a very slow-motion train wreck, due to the snap-back potential. Car companies are spending a lot of money on EVs and it could turn out to be mainly wasted, since market penetration may never exceed 10-20%. The IEA projects about an 11% market share by 2030 (good luck with that). Market share for EVs in 2017 was about 1.15% of cars sold. Like the "overall solution" proposed by governments, the car companies are locking into an approach that is not, and probably cannot be, cost-effective. This will lead to continued pressure on margins. All of the emission analysis in this article assumes a broad push to lower profit vehicles (higher mileage), at the same time as huge investments are being required. Car companies got into trouble when the market zigged from high mileage small cars back to our favorite trucks and SUVs. The same is likely to happen with EVs.


EVs will continue to have a troubled future. Tesla will be the first to feel the practical problems yet to be solved, but the other auto companies aren't far behind. The utilities face a different set of problems, but the need to raise huge sums of capital seems to be the initial hurdle. I am long Ford and GM because they provide a good dividend stream at the moment. For the longer term, I will be closing down those positions and replacing them with companies that have a more certain future.

Disclosure: I am/we are long GM, SO, ED, F. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.