New Numbers Are In and EVs Are Cleaner Than Ever

, senior engineer, Clean Vehicles | May 31, 2017, 4:40 pm EDT
Bookmark and Share

This post is a part of a series on Clean Energy Momentum

One of the most common questions I’m asked about electric cars is, “how clean are they?”

Five years ago, UCS answered this question, publishing its first look at the global warming emissions from electric vehicles (EVs) in our ‘State of Charge’ report.  In early 2017, the US EPA updated their data on emissions from electricity generation, now capturing power plant emissions through the end of 2014. How does this new data change our assessment of EVs?

For over 70 percent of Americans, driving an EV results in fewer emissions than even a 50 MPG gasoline vehicle.

We now find the overall global warming emissions from using an EV is significantly lower for most of the US. Several regions of the country showed significant decreases in emissions, as compared to our first EV emissions assessment.

When compared to our initial report on EV global warming emissions, the changes are impressive. That report used 2009 power plant data (the most current available in 2012) and placed only 9 of 26 regions in the ‘best’ category. Now 19 regions are in the best category with only 2 in ‘good’ regions. For example, the Northern Midwest region that includes Minnesota and Iowa improved from 39 MPG equivalent to 54 MPG and Eastern Wisconsin also jumped from ‘good’ at 40 MPG to our ‘best’ rating with emissions equal to 52 MPG gasoline cars.

Global warming emissions from electricity generation have fallen in since 2009 in many parts of the US, making EVs even cleaner. Check out the changes by region in the slider above. 

Based on where EVs have been bought to-date, the average EV in the US now produces emissions equivalent to a hypothetical gasoline car achieving 73 MPG.

Nearly half of the EVs sold to date have gone to California, where the average EV produces global warming emissions equal to a 95 MPG gasoline car. The next 5 states for EV sales (Georgia, Washington, New York, Florida, and Texas) account for 20 percent of US EV sales and are regions that have emissions ratings of 50 MPG or better.

Manufacturing emissions are important, but much less of a factor than fuel emissions.

The emissions estimates presented above compare the use of an EV compared to using a gasoline vehicle. However, there are also emissions associated with the production of these cars, and in general making EVs produces more emissions than a comparable gasoline car. We studied this issue in our “Cleaner Cars From Cradle to Grave” report in 2015 and found that the extra emissions from making an 80-mile range EV (compared to a similar gasoline car) are about 15% higher. However, this extra emissions ‘debt’ is quickly recovered by the savings that accrue while using the electric vehicle.

How quickly the emissions are recovered depends on where the car is charged, but for an EV the size of the Nissan LEAF, we found that break-even point occurs after 6 to 13 months of use (depending on electric grid region), well shorter than the likely lifespan of the car.

Choosing an electric car over an inefficient gasoline model is one of the most influential decisions a household can make to reduce emissions

For the average American, transportation makes up about a third of all household global warming emissions. And compared to some other sources of emissions, we have a great deal of control over how efficient a vehicle we choose. The average new gasoline vehicle in the US is rated at 25 MPG. On average, driving an EV (at 73 MPG equivalent emissions) would produce global warming emissions at less than half of the rate of the average new vehicle.

If you’re curious about how clean specific EVs would be where you live, check out our EV tool here. It’s recently been updated with our newest estimates of EV emissions, and we’ve also added many new EV models. If you are interested in the most efficient (and lowest emission) EV models, check out the Hyundai Ioniq BEV, Chevy Bolt, and BMW i3 BEV models.

Changes since our last report include generation, fuel production, and transmission efficiency.

Our initial assessment comparing gasoline vehicle emissions to those from electric vehicles were detailed in our 2012 State of Charge report. That report relied on the best data available at the time. This included estimates of power plant emissions and transmission losses from 2009 and also included the most recent estimates of ‘upstream’ emissions (such as coal mining and oil refining).

While we used the same analysis method as both the State of Charge and Cleaner Cars From Cradle to Grave  reports to generate these new emission estimates, the input data has changed.

The EPA estimates of power plant emissions in their eGRID database have been updated from 2009 data to 2014 data. In many cases, the emissions from power plants decreased, often due to reductions in coal-fired power and increases in renewable generation. However, some regions did show an increase. For example, in the Pacific Northwest, hydroelectric power output was reduced and fossil fuel plants supplied additional power.

The eGRID data also includes an updated method for calculating the losses attributed to the transmission and distribution of electric power from generators to the end user. This loss estimate is significantly lower than previous estimates, and therefore lowers the emissions attributed to EVs.

Finally, we also updated the estimates of emissions from ‘upstream’ sources like fuel extraction and refining. We used the most recent version of the GREET model from Argonne National Laboratory to estimate these emissions.

Most regions showed a decrease in emissions from electricity generation and distribution from 2009 to 2014. Red triangles indicate the total change in global warming emissions due to changes in generation sources, upstream emissions from fuel production, and losses in transmission and distribution of electricity from power plants to the end user.


Posted in: Vehicles Tags: , , , ,

Support from UCS members make work like this possible. Will you join us? Help UCS advance independent science for a healthy environment and a safer world.

Show Comments

Comment Policy

UCS welcomes comments that foster civil conversation and debate. To help maintain a healthy, respectful discussion, please focus comments on the issues, topics, and facts at hand, and refrain from personal attacks. Posts that are commercial, self-promotional, obscene, rude, or disruptive will be removed.

Please note that comments are open for two weeks following each blog post. UCS respects your privacy and will not display, lend, or sell your email address for any reason.

  • BetelgeuseOrion

    what are you guys even comparing? vehicle emissions? or fuel generation emissions? you dont cross compare because that makes no sense.

    EVs = 0 emissions at vehicle level
    ICE = Lots of emissions at vehicle level

    EV’s = variable emissions from fuel generation
    ICE = Lots of emissions from fuel generation

  • These ratios have nothing to do with the efficiency or cleanliness of electric vehicles. There is no measurement of average EV operation times, the ratios of EV fleets to ICE, comparison of the effect of anti-emission systems to the increase of demand in support of EV fleets, conversion and loss rates for generation / transmission / stepping / regulation / environment potential / storage systems and materials. Why not calculate the average loss for a standard transmission system, from source material to battery, then calculate the amount of material consumption and compare efficiency to internal combustion? The number isn’t anywhere near 50 or 73 MPG, probably not even 20! This report is fraudulent.

    • No data to support your claims, therefore you are being fraudulent.

      • The data is written in physics books, so return to college and learn to read one.

    • We’ve edited several of the comments to remove language that violates our comment policy. We welcome robust discussion, but please keep it civil and refrain from personal attacks. Thank you!

    • BetelgeuseOrion

      and the 25MPG rating of mid sized cars do not include

      Seismic and ground radar survey / Drilling / completions / transport /refining / pipeline / tankers / pumping / etc

      if you want to include manufacturing emissions from the batteries etc, do the same for gasoline vehicles.

      • My argument focuses specifically on fuel efficiency of the vehicle (the subject of the article), not manufacturing processes. The refueling process for an EV involves consumption and exhaust at the source, conversion, transmission, regulation, and storage. I don’t need “Seismic and ground radar survey / Drilling / completions / transport /refining / pipeline / tankers / pumping / etc” to fuel my vehicle, however it isn’t possible to recharge an EV without conversion and transmission, therefore we cannot compare efficiency without measuring conversion and transmission losses. “We’ll eventually add battery packs at our hundreds of 3+ megawatt-per-hour supercharging stations” and “The grid is getting cleaner, so give us a chance to fix these phasing and resistance problems” are not valid counter-arguments. Even laymen can see that it will never be possible to support a fleet of EVs with renewables which can’t even meet basic residential requirements, even if you blanket every functional space with panels.

  • Question

    Is it safe to assume the batteries of these EVs are lithium based? If so, what is the impact on the environment to mine this resource and the longevity of it?

    • BetelgeuseOrion

      if its safe to assume that gasoline comes from crude oil or bitumen, then what is the impact on the environment to drill for that resource?

  • Will Toor

    While I understand why you use the grid regions and 2014 data for this analysis, it can lead to some misleading conclusions. In Colorado, for example, there is very little power imported or exported, but the state is lumped in with much dirtier power in Wyoming and Nebraska, and each time this report comes out we get stories that claim that Colorado is the worst place in the country to have an EV. If you analyze the power mix in Xcel service territory, where the vast majority of EVs in Colorado are registered, and you use 2016 data, the equivalent MPG would be 47, not the 38 quoted in the UCS study (we have just done this analysis for the City of Denver and the Colorado Regional Air Quality Council, and will be releasing the study in July). We also found that with the adopted integrated resource plan, by 2025 the equivalent MPG will be 72 mpg. I harp on this because every time this study is updated we end up doing significant damage control with Colorado policymakers because of the inaccurate picture it paints within Colorado.- Will Toor, Southwest Energy Efficiency Project

  • Matt Uyttendaele

    I’m curious where the efficiency numbers for different EVs come from? For example it looks like the Chevy Bolt is 25% more efficient that the Tesla Model S.

    • The efficiency data is from and is based on the EPA fuel economy test.

      • Matt Uyttendaele

        Thanks for the reply! Just confirming – looks like you used the highway numbers vs city or combined city/highway.

    • There are never any actual calculations or detailed statistics because they fabricate it. This is called fraud.

  • Too many “low-information” hit-piece writers and trolls out there continue to restate the “long tailpipe” myth, and I cite UCS to disprove their claims. I’m glad to see the update showing that the numbers are even better. Please keep it up!

    • We’re tired of the “no burning liquids = no emissions” myth. Even if you don’t consider material manufacturing, EVs are not more efficient than vehicles with internal combustion engines when recharged over any grid, and there will never be any possibility of powering a full fleet with local solar arrays.

      • RTFA. It is a study encompassing the entire cost of ownership, manufacturing, and disposal of these types of cars.

        If you truly care about emissions, advocate to phase out fossil-fuel powered plants in favor of renewables like solar and wind and using stationary storage like pumped-storage hydroelectric and batteries to load balance (that’s how solar can power all of the grid). The above shows most of the country is trending that way. No one is claiming “no emissions” as of 2014 (the data source the UCS uses). It’s less emissions now than with ICE, less and less every year, ultimately tapering to negligible.

      • RAFPB. This is an attempt to convince laymen that we’re on the threshold of a ‘clean energy revolution,’ while there is no possibility of supporting a productive electric fleet with solar arrays due to transmission losses, low conversion rates, and lack of surface area. It’s not less than anything, you just can’t read physics.

      • Your claim is completely refuted years ago. The amount of area needed to power all human electricity consumption is just a smidgen of the Earth. They’ve already worked out the math for you:

      • You’re obviously delusional: “Centuries of empirical research has been refuted! –Insert reference to two-paragraph article–” Go and read a BOOK. This article is fraud.

      • Today’s solar panels, and other technologies, didn’t exist centuries ago, but I can’t stop you from living like the Amish and imprisoning your own mind in something much worse like your own echo chamber.

      • Physical constants will never change after any number of centuries. I’ll refer you back to the other thread where I mention “conversion and loss rates for generation, transmission (cable resistance), stepping, regulation, environment potential, storage systems and materials.” What you can stop is yourself from attempting to counter substantiated physics arguments with distractions and false calculations. The public is beginning to see that the united states exists only to promote fraud, so please try actually reading a book before your next reply.

      • No post of yours here states or cites an actual calculation. A laundry list is not physics nor is a valid argument. The calculations in the article I cite, as well as the other articles that one links to:

        are substantiated by and consistent with the known laws of physics, the parameters of our solar system, and the constants of the universe. You offer no proof otherwise (again a list is not proof). Please actually try reading all of them before attempting to reply.

        “the united states exists only to promote fraud”? I thought people like you said China was promoting hoaxes.

      • None of your statistics have anything to do with calculation of conversion and loss rates. If they were consistent with ‘the known laws of physics’ then you would be able to answer two simple questions: 1. Based on standard configurations for electrical generation and distribution systems, what is the range of efficiency ratings for EVs which are recharged over any grid, and how do those ratings compare to the efficiency of internal combustion engines which are powered by liquid fuels? 2. What is the average emission rate for EVs which are recharged over the grid when transmission and conversion losses are considered? The efficiency is lower than gasoline, and there won’t be any possibility of supporting multi-megawatt-per-hour supercharging stations with local solar arrays. You can’t answer these questions because you do nothing more than regurgitate fabricated metrics. Go and read physics book as I’ve instructed you to, you illiterate cretin.

  • Oil4AsphaltOnly

    Thank you for the updated report! Especially highlighting the changes from the previous report.

  • Chuck Swackhammer

    You didn’t include solar powered homes charging their electric cars!

    • Hi Chuck- The emissions data shown here is an estimate of average power plant emissions in a region from driving on electricity. You are correct that home solar is not reflected in these estimates. Other factors would also impact emissions at a household level, such as green power purchase options and local utilities that have more renewable energy on their system. However, this report is designed to answer the question of how regional electricity generation differences impact average EV emissions.

      • Chuck Swackhammer

        Thank you for clearing that up. People need to know that they can drive on sunshine even in areas they wouldn’t expect. The day I energized my system and put it on the grid it was snowing and I was already producing more energy than I needed.