Roskill: predominance of lithium production shifting to mineral vs brine, driven by volume requirements of EVs
ULEMCo working on hydrogen-fueled marine engines

Study of 5-year TCO suggests ongoing challenge to unsubsidized BEV cost-competitiveness

A team at Arizona State University has analyzed the five-year Total Cost of Ownership (TCO) for representative electric, hybrid, and conventional vehicles—the Nissan Leaf (BEV), Toyota Prius (HEV), and Toyota Corolla (ICEV)—in 14 US cities from 2011 to 2015.

The results, reported in a paper in the journal Energy Policy, show spatial variation due to differences in state and local policies, fuel prices, insurance and maintenance costs, depreciation rates, and vehicle miles traveled. Despite those differences, in nearly all cities, the BEV’s higher purchase price and rapid depreciation outweighed its fuel savings.

Sensitivity analyses highlighted the impact of key parameters and showed that both federal and state incentives were necessary for BEVs to be cost-competitive.

Future BEV cost competitiveness may improve if innovation and scaling lead to significantly reduced BEV purchase prices, but our analysis suggests that it will be challenging for BEVs to achieve unsubsidized cost competitiveness except in the most optimistic scenarios.

—Breetz and Salon

To estimate ownership costs as realistically as possible, Breetz and Salon used a five-year ownership period (Fiscal Years 2011–2015) and city-specific data on vehicle mileage, fuel prices, insurance costs, maintenance and repair costs, resale values, and taxes, fees, and subsidies.

They also conducted sensitivity analyses exploring the impact of fuel prices, discount rates, depreciation rates, length of ownership, and driving distances—and what it would take for the Leaf to achieve cost-competitiveness without federal and state subsidies.

We find that although ownership costs varied considerably across cities, the Leaf cost substantially more than the Corolla in all cities and more than the Prius in all but one city. A principal reason is that the Leaf depreciated faster than the gasoline vehicles, losing more in resale value than it gained in fuel savings in the first five years. In addition, the Leaf’s higher purchase price resulted in higher sales tax, ad valorem taxes, and insurance costs. Sensitivity analyses demonstrate that an owner may save money with the Leaf compared with the Corolla or Prius, especially if they have access to free or reduced-rate charging. Government incentive programs were still necessary, however, for the Leaf to achieve cost competitiveness.

—Breetz and Salon


Five-year TCO by city and car. Breetz and Salon.


  • Hanna L. Breetz, Deborah Salon (2018) “Do electric vehicles need subsidies? Ownership costs for conventional, hybrid, and electric vehicles in 14 U.S. cities,” Energy Policy, Volume 120, Pages 238-249 doi: 10.1016/j.enpol.2018.05.038



And long range BEVs?

Lets not be silly


Agreed, another contrived study to take advantage of the rapid innovation in the BEV field and cast EV’s in a negative light!

Short range EV’s are like pre-Prius hybrids, under powered, overpriced and have low resale value. Tesla’s on the other hand, have amongst the highest resale value of any cars made. One of the major reasons other EV’s have low resale value is they are not Tesla’s!

The authors of this study should be ashamed of themselves!

Brent Jatko

For the portion of the market that is able to charge at home, the ability to begin each day with a "full tank" is a non-monetary reward which also needs to be factored in.

But I basically agree with Davemart and AndrewsRobert's objections; this study seemed to select the EV (Nissan Leaf) with the worst depreciation in order to skew the results.


The solution, then, is to buy a used leaf.
It would be interesting to try some more modern cars in it, from say 2014 - 2018.



You seem to have missed that Andrew and I have diametrically opposed views.

Any comment in which Tesla’ and ‘sensible economics’ are cojoined is mistaken, IMO.


Sounds like this is to convince dolt’s and dotards. They consider unsubsidized BEVs but they include depreciation. They probably failed to adjust depreciation by the subsidies which would completely invalidate their findings.

They also probably did not analyze the case of a driver who owns a solar PV array even though surveys have shown that is about half of BEV owners.


Same maintenance cost for Corolla and Leaf?


For unknown reasons, the evaluators picked one of the worst BEV from Nissan against the best ICEV/PHEV/HEV from Toyota.

Picking the best more costly BEV could have been worst, specially in USA where fuel is very cheap.

Most BEVs are not yet competitive and may not be before much better (4X and 5+X) lower cost longer lasting quicker charge batteries are available.


When Battery Tech catches up and becomes less costly, EVs will be more competitive with ICEVs. However, there are reasons to buy an EV other than cost. They are cleaner, quiet, easier to drive, less polluting and more fun...well worth the extra money.



Good comments,
the no gas stations/full "tank" every morning works.


The auxiliary tank that always starts full in the morning is also great about PHEVs.  I'm still driving on a tank I filled in March.


Currently, for people with Level I or II home charging facilities, a BMW style PHEV, with large batteries and very small ICE generator, is probably one of the best compromise to compete with ICEVs, reduce fuel consumption, pollution and GHGs.

As batteries performance increase and cost per kWh is reduced, the ICE generator could be reduced and eventually eliminated by 2030 or so?


I wouldn't plan for 100% electrification of most LDVs until at least 2050.  But if you can electrify 80% of the mileage and use e.g. biofuels for the rest, you can get to 100% decarbonization quite a bit sooner.

Bob Niland

The primary financial number that matters for any vehicle use (owned, leased, rented, uber, ...), of whatever motive power, is
TCO/mi (or /km if you prefer)
and we don't really get that in this report. Due to politics, TCO/mi is, of course, presently widely varied by jurisdiction for EV and hybrids. But some national model could be proposed.

The ICE car business has never been much inclined to discuss TCO/mi. But if some dissident brands want to break that taboo, there is a rough benchmark that consumers can compare to, and that's the annual IRS mileage deduction.

Concur with the critiques of the models chosen for the report. Also, personally, I buy-new and drive-into-ground, so re-sale value is irrelevant to me, but any standardized TCO/mi model would have to make certain assumptions that don't work for everyone.


Something close to 100% electrification may not be reached during this century due to ships, trains, large aircraft, heavy machinery, war machines etc unless FCs evolve enough to compete favorably with ICEs.

The first generation fuel cell trains will be in operation in Germany by the end of the year. Many specialized boats/ships, trucks and buses will follow.


...and cobalt prices are exploding... at the same time as child labour under terrible conditions in Congo is increasing. Not a good future for EVs.


I'm one of the authors of the publication. And -- wow -- I'm pretty startled by the vitriol of some of the comments. I encourage everyone to read the actual paper, because it addresses many of the issues that are raised in the comments.

For what it's worth, the paper was not intended to make EVs look bad. Our goal was to understand spatial variation in ownership costs -- that is, how EV ownership costs varied across major US cities depending on factors such as tax policies, driving patterns, relative gasoline/electricity prices, and vehicle depreciation rates. We wanted it to be an empirical analysis based on real-world data, unlike most of the TCO literature, which is based on simulation models. That meant we had to use an older vehicle because at the time of data collection, the 2011 Nissan Leaf was the only EV where we could get a full five fiscal years of data (not just about the cars, but also about policies, fuel prices, etc.). And because we were interested in speaking to debates about mass adoption, we chose the most popular models of each type of car (ICEV, HEV, BEV). We ran tons of sensitivity analyses and counterfactual scenario analyses to explore the robustness of the results. It's a thorough paper that has a lot of depth in it that's not apparent just in the summary above.

We expected to find that, for the 'typical' vehicle owner, EVs make financial sense in some cities but not others depending on local policies, prices, and driving distances. That wasn't the result we got. But that's the way research works; it doesn't always give you what you 'want' to hear. That's often where you learn the most. Doing this analysis demonstrated several important point about the structure of costs. This includes the importance of variable depreciation rates and state-level policy, both of which are ignored in most TCO studies. We also showed how much variation there is across cities, which may seem obvious but hadn't previously been studied or quantified (the vast majority of the TCO literature uses nationally-representative parameter values).

Anyways, I just wanted to provide a bit more explanation and background. If you read the paper, you'll see it's a rigorous analysis with a lot of depth and nuance, not a hit job on EVs. Please feel free to contact me if you'd like help getting a copy of the paper.


Do please send me a copy, Hanna.  My e-mail is on the sidebar of my blog,

Your 5-year data cutoff would also eliminate the PHEV I drive, so I can't blame you for not rating it or its closest cousin.  But my car turned 5 a couple months ago, so I look forward to seeing them rated in the future.


Hmph.  Still no paper, and no obvious way to find a contact address.


Sorry Engineer-poet -- didn't actually see your email address in your sidebar. My full name is in the blog post as well as being my username. Easy to pull me up in quick web search.

The comments to this entry are closed.