Three Facts About EVs and Multi-Vehicle Households – Energy Institute at Haas

Energy Institute Blog
Research that Informs Business and Public Policy
Within-household substitution between vehicles could undermine the environmental benefits of EVs.
I got interested in how multi-vehicle households use EVs, because — from a carbon emissions perspective– we care not only about how many EVs are sold, but also about how they are used. After all, it isn’t the manufacturing of EVs that gives them their environmental edge, it’s their potential to offset driving gasoline and diesel-powered vehicles. 
The driving behavior of multi-vehicle households is particularly important to understand because these households have so many options with regard to the combination of vehicles they choose to have, as well as the scope for substituting across vehicles when making trips. I’m worried, in particular, that this within-household substitution could undermine the environmental benefits of EVs. This wouldn’t matter in a future scenario with 100% EVs, but it could matter a lot during a long transition period.
So here are some facts based on the 2017 National Highway Transportation Survey. I’ve posted details about all calculations as an Energy Institute working paper here. These are the most recently available nationally representative data. But they are already several years old, and EV market offerings are changing fast, so I will also discuss later how I think these patterns will change over time. 
Fact 1: 90% of U.S. households with an EV also have some other vehicle.
As the figure below illustrates, only 10% of U.S. households with an EV are single-vehicle households. In contrast, 37% of all U.S. households are single-vehicle households. Thus, households with an EV are almost four times less likely to be a single-vehicle household.
In the U.S., it is even common for a household to have more vehicles than drivers. Among U.S. households with an EV, 36% have more vehicles than drivers, compared to 24% for other U.S. households. 
This all points to there being considerable scope for within-household substitution. Just because a household has an EV doesn’t mean that will be the vehicle it takes on family vacations, for example.
Fact 2: 60% of U.S. households with an EV also have a non-electric SUV, truck, or minivan.
The table below describes the other vehicles in U.S. households with an EV.  Cars and SUVs are the most common, but many EV households also have trucks and minivans. Overall, 60% of households with an EV have a “large” non-electric vehicle, i.e., an SUV, a truck, or a minivan.
table 1
This pattern makes a lot of sense. These larger vehicles provide differentiation with regard to exterior and interior dimensions, seating capacity, cargo area, and other attributes. In recent research, economists James Archsmith, Ken Gillingham, Chris Knittel, and Dave Rapson find evidence that households substitute between attributes when deciding which vehicle to purchase. For example, a household with one fuel-efficient vehicle may be more likely to purchase a second vehicle that is less fuel-efficient. This substitution between attributes likely plays a particularly important role with EVs, and may help explain why EVs are so popular in multi-vehicle households.  
This differentiation is convenient for the household, but it also has implications for fuel economy and gasoline consumption. These larger vehicles in households with EVs have an average fuel economy of 18.8 miles-per-gallon, which is lower than the average fuel economy for the U.S. vehicle stock.
Fact 3: 66% of U.S. households with an EV have a non-electric vehicle that is driven more.
Finally, two-thirds of households with an EV have some other non-electric vehicle that they drive more miles per year. This is somewhat surprising. EVs cost less to drive per mile than gasoline-powered vehicles, so there is a financial incentive for households to use EVs intensively. It isn’t clear why, but there seems to be something about the attributes of EVs that tends to work in the opposite direction.
Regardless of the exact explanation, this substitution of driving toward non-electric vehicles reduces the environmental benefits of EVs and provides additional context for previous studies (here and here) that have tended to find that EVs are driven less than other types of vehicles.
empty road
New Generation
This all points to the significance of the newer generation of EVs which include not only cars, but SUVs, trucks, and even this three-wheeled RV. The number of available EV models has expanded considerably and this gives households more flexibility and less need to combine EVs with other non-electric vehicles.
Range is a good example. The earlier EVs had less range than practically any of the models available today. The first generation Nissan Leaf, for example, had a range of less than 80 miles, while the current day Leaf has a 150+ mile range, almost twice the original version.
In the language of a new paper by Stephen Holland, Erin Mansur, and Andrew Yates this means that EVs are becoming closer substitutes to non-electric vehicles, a trend that bodes well for EVs as a climate solution.
Nevertheless, I still think future research needs to think hard about multi-vehicle households. Over the next decade there are going to be millions of households with both an EV and a non-electric vehicle, so we need to understand better both how households decide which vehicles to purchase and how they decide which vehicles to drive.
Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.
Suggested citation: Davis, Lucas. “Three Facts about EVs and Multi-Vehicle Households” Energy Institute Blog, UC Berkeley, September 20, 2021,
For more details see “Electric Vehicles in Multi-Vehicle Households”, by Lucas Davis, Energy Institute Working Paper #322.
energy efficiency, gasoline, transportation
Lucas Davis is the Jeffrey A. Jacobs Distinguished Professor in Business and Technology at the Haas School of Business at the University of California, Berkeley. He is a Faculty Affiliate at the Energy Institute at Haas, a coeditor at the American Economic Journal: Economic Policy, and a Research Associate at the National Bureau of Economic Research. He received a BA from Amherst College and a PhD in Economics from the University of Wisconsin. His research focuses on energy and environmental markets, and in particular, on electricity and natural gas regulation, pricing in competitive and non-competitive markets, and the economic and business impacts of environmental policy.
Lucas, alas this is one post where I think you blew it. EVs are almost always the *preferred* vehicle to drive, but in 2016 most were compliance cars and low range meant they were constrained to local trips, with the ICE vehicle used for long trips and other uses (dump run etc). The long trip scenario is fast disappearing, and I’m pretty sure an updated study will show very different results. (Owner of a 2015 Ford Focus EV and 1999 4Runner.)
One important factor left out of the article and the discussion so far is EV charging speed. One reason someone might use an EV only for short trips is not only range, but charging speed. That is because unlike the Tesla, many non-Tesla EVs take much longer than Tesla’s to charge.
For example, one can drive from the S.F. Bay Area to San Diego in a Tesla in a time comparable to a gas-powered car. However, we tried that in our Chevy Bolt and had to stop in L.A. to sleep and charge overnight because it would have taken too long to fully charge for the last leg of the trip. We love our Bolt. But our next EV will be one that has a good range and can handle sufficient kW to fast charge at the supercharging stations being installed for non-Tesla vehicles. Electrify America has such charging stations that charge at rates of 50 – 350 kW. Many EV drivers don’t know this because they haven’t tried taking them on long trips. Probably for this reason, EV manufacturers also don’t advertise their low charging speeds in their specifications.
In the future, in our 2 person household, we could have one EV for short trips and one for long trips. Note also that some organizations classify Chevy Volt and other plug-in hybrids as EVs because they have the capability to be only used as electric. However, Volt owners like us may be skewing EV data because Volts allow us to use them as EVs for short trips, and use them for long trips. But the long trips use mostly gas because Volts only have short EV range.
You have both a Volt and a Bolt? The Bolt must not have fast charging capability because you can charge it in about 1.5 hours, not overnight as you did in L.A. I’ve taken mine over the Sierras to Mammoth and Mono Lake and not had any trouble with the time.
P.S. The Bolt has 239 (or 259 for later models) Certain Tesla models are about the same I believe.
As pointed out by Jim Lazar and Carl Wurtz, the characteristics of EVs have changed substantially since 2016 when the survey was conducted (Apr 2016-Apr 2017). Whereas almost everything was a coupe, Car and Driver devoted an entire issue to “EV of the Year” which as populated by all sorts of SUVs. ( And EVs are being touted for other benefits. Here’s the Ford F-150 Lightning ad running during the Olympics that touts its ability to serve multiple roles.
A key issue in the presentation here is that the comparisons are made between EV owners and the general population. The fact is that EV owners in 2016 (and still true today) were wealthier than the average household and almost certainly owned more vehicles. In additions, EVs are much more likely to be “sold” through a lease, which creates added barriers that further narrow ownership to higher income households (which makes me think that leasing through either local governments or utilities instead might be a useful means of reaching lower income households.) Further, lower income households are much less likely to buy new cars and there have been few EVs available to date.
The answer to the what type of research should be done is to focus on how use changes with these changes in both technology and ownership demographics. A static analysis of what it was like 5 years ago, reporting what, as one commenter said, we already knew doesn’t help our current understanding much. It’s an important baseline of a much larger study, but not truly informative for our understanding what is likely in the future.
Based on the data I’m linking to, I’m guessing that the majority of electric vehicles on the road in the US today are newer than 2017. That’s the end of the data set reviewed in the article you’ve shared(though the article just came out).
The rapid growth of EV use should make us cautious about interpreting trends.
I’m in a one vehicle home with an EV, bought this May. I recognize that I’m anecdotal. I also recognize that the data interpreted here, while only a few years old, is out of date.
Relying on the 2017 National Highway Transportation Survey is the Achilles heel of this blog posting. At that time, there were a few long-range EVs on the road, and a whole bunch of early Leaf vehicles with less than 100 miles of range. Today the new EV sale mix is completely dominated by vehicles with 200+ miles of range: Tesla, Chevy Bolt, Kia Niro, Nissan Leaf, Audi E-Tron, Ford Mustang, Hyundai Kona, and more.
The 2017 NHT Survey did a reasonable job examining EV usage in that era. A 2011 Leaf, our previous EV, could not make it to Karen’s sister’s house for a visit. The new Niro (or Leaf, or Kona, or Bolt) can do so easily.
So, in 2017, with a 2011 Leaf and a 2002 Odyssey, we would have taken the ICE vehicle (a Honda Odyssey). Today we would take the Niro. It’s not a matter of range anxiety, simply one of single-trip range. In the three years we’ve had the Niro, we’ve done exactly three trips that required advance planning and mid-trip charges — or using the ICE. As compared with every week with the 2011 Leaf.
And, indeed, since acquiring the Niro, the annual mileage of the Odyssey has dropped from about 8,000 miles/year down to about 4,000 miles/year, while the mileage of the EV has increased from about 6,000 miles/year on the Leaf to more like 9,000 miles/year on the Niro. But we also transitioned from one retiree to two retirees in that period, and we bought recumbent trikes (which only fit in the Odyssey) so there are some variables at work beyond the fleet upgrade.
It will be useful to see what the NEXT NHTS shows.
“Fact 3: 66% of U.S. households with an EV have a non-electric vehicle that is driven more.”
A few observations from someone living in a two-car household with one EV.
There are periods where our non-EV gets more usage than our EV, even for short trips when the EV is fully charged. The explanation is simple: for most EV drivers living in a single-family household, the charger is behind their house – either installed in a garage, a carport, or under an eave of the house. If the driveway only accomodates one vehicle at a time, the non-EV gets parked in the driveway behind the EV. It gets used more often.
My EV is a 2011 LEAF, with less than 50 miles of range left (note: in 2011 its range was over 100 miles). So it gets charged often, and ends up in front of our other car. Our other car is a Ford Escape hybrid, which for an SUV gets an entirely-respectable 30 mi/gal.
Many new generation EVs (among them the Tesla Model 3, Chevy Volt) come with options for over 200 miles of range, so I foresee range anxiety being a problem solved by lower prices for battery capacity. Turo ( now offers an “AirBnb for cars”, which, together with lower prices of rental cars, could one day replace the need to own a longer-range ICV for road trips.
An interesting observation. This is one illustration of how the results of this retrospective study using data from 5 years ago probably isn’t a good basis for making prognostications about future EV use.
The statement: “this substitution of driving toward non-electric vehicles reduces the environmental benefits of EVs” is only true if you initially assume, for some reason, that an EV is going to substitute for the average existing vehicle in the US fleet. Nobody should be making that assumption, at least not yet. Of course EVs are driven less than average and less than other vehicles. If a two-car household has one driver with a 100-mile round-trip everyday and another with a 50-mile roundtrip everyday, the EV is going to be making the shorter trip, due to the limited average range of the existing stock of EVs and existing limitations in charging infrastructure. Only with the coming expansion of long-range EV availability is this going to begin to change. And it’s going to take even longer for it to be reasonable to think that people are going to take their one EV “on family vacation” instead of an ICE. This does NOT mean, however, that EVs don’t have environmental benefits. Please be mindful that if you make a statement that some fact or behavior “reduces the environmental benefits of EVs” that the statement will likely be taken out of context or misinterpreted by people who do not want to see EVs succeed. I’m not saying to hide or gloss over the limitations of EVs, which are still real and a concern for many drivers, but to be careful of context and to make fair comparisons and claims. Even an EV driven 6 or 8 thousand miles a year can have enormous environmental benefits (and yes, that is dependent on the carbon intensity of the grid from which they charge), so let’s not give skeptics reasons to thwart the move to electrified transportation that must be a critical component of decarbonization.
Common sense. Doesn’t require a PhD. Any high school kid could have thought this thru and validated with a quick ‘unscientific’ survey of those with EVs.
I have small solar panels that create about one gallon of gasoline equivalent energy per day fuel when comparing miles driven in a gasoline powered car versus my Tesla. Here is what the panels look like. While in Austin the energy going into the Tesla is purely solar energy so there is no load on the electrical grid nor is there any gasoline burned.
Visit Our Website
Join Our Email List
Donate Today
Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to the author and the Energy Institute with appropriate and specific direction to the original content.


Leave a Reply

Your email address will not be published. Required fields are marked *