Skip to main content

What CityLab Gets Wrong about EVs

How one error keeps tripping up analysis of electric vehicles

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


Electric vehicles (EV) tend to ignite passionate debate. For anyone who has worked on the topic (disclaimer: I have), it seems you can’t even bring up the topic without being put into a “pro-EV” or “anti-EV” camp. The same is true for hydrogen-powered fuel cell vehicles – but let’s stick with EVs for today’s discussion.

In late June, Eric Jaffe published a post in CityLab [and a follow-up post] that discussed the not-so-surprising fact that EVs are only as clean as their upstream energy source. His post was based on recent analysis by Stephen Holland from the University of North Carolina-Greensboro, which did a good job of accounting for differing viewpoints on vehicle electrification, but missed a rather salient point - the zero-sum fallacy.

In other words, just because one solution might look quite appealing (e.g. solar power) doesn’t mean it has to be the only one we use. We can have both solar and wind power, better public transit and increased fuel economy, and we can fight poverty and climate change at the same time. Yet, we’re human beings, and human beings want one solution, and they want it today. However, the answer is several solutions, over time. It’s not as “sexy”, but it’s the only sustainable path forward.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


According to Holland’s research, EVs are only as green as their power source. This is true, but perhaps not really “news” per se. Electricity is the fuel source for EVs, and if the power grid isn’t decarbonized, then EVs are not truly zero-carbon. For a significant portion of the US, today’s power sector is carbon-intensive, and EVs that charge in these states are currently far from being zero-carbon.

Estimated damages for gas (left) and electric (right) cars by U.S. county; the damages range from roughly 1 to 5 cent(s) per mile on each side, green to red. (Holland et al, 2015, NBER)

Estimated damages for gas (left) and electric (right) cars by U.S. county; the damages range from roughly 1 to 5 cent(s) per mile on each side, green to red. (Holland et al, 2015, NBER)

But, this analysis takes a short-term view – and does not consider the evolving (and according to policies and trends, decarbonizing) energy sector from a systems perspective. In other words, the power sector is first in line to be decarbonized, and plans are well underway in the US, EU, and elsewhere. Why? Because the power sector already has the lowest abatement cost of any energy end-use sector i.e. it’s currently cheaper to replace coal with solar panels, than to replace gasoline cars with EVs. The long-term implication? The fuel that EVs use (electricity) is becoming less carbon-intensive over time.

While Jaffe’s post at CityLab acknowledges that we need to compare apples-to-apples (lifecycle impacts of conventional vehicles vs. electric vehicles), it still ends up falling into the biggest trap of all when he concludes by saying, “And if it’s really all about the future, [Holland] says, let’s subsidize car research instead of car purchases.” This zero-sum fallacy is unfortunate since the author goes to great lengths to avoid it, yet ends on that note. Let me explain what’s missing.

If we want to decarbonize the transport sector, in line with climate change goals (not to mention for the sake of other co-benefits such as local air pollutants) then you need to start ramping up sales of zero emission vehicles like EVs today. This ramp-up will allow you to hit interim greenhouse reduction targets and position you to switch completely over to EVs in the long-term, say 2050 or so.

The suggestion that we should be subsidizing conventional car research to achieve increased fuel economy instead of supporting EV deployment is a fairly fatuous proposition. Governments have been subsidizing car research for over a century, in many, many forms, and bailouts. And while gasoline- and diesel-powered cars need to become cleaner to play their role in decarbonizing the transport sector (and are the low-hanging fruit in many ways, especially through hybridization), we need zero-carbon vehicles such as EVs to start ramping up now in parallel.

It would be nice of course if we could just switch over to EVs once the grid is fully decarbonized, but here’s why that doesn’t work:

  • First, you would want economies of scale in place to have driven down the upfront purchase price of EVs to decrease the total cost of such a transition

  • Second, you want infrastructure in place, both in terms of physical installations, but also in terms of best practices of business models and technical standardization (you can see this in California’s efforts to install zero emission vehicle (ZEV) infrastructure (and Tesla’s for that matter) to spur development and adoption)

  • Third, given that a car’s lifetime might be somewhere around 12 years, a car bought in a place with a decarbonizing power sector actually reaps benefits increasingly over time. It’s becomes “cleaner over time” if you will. But, if you don’t put this EV on the road, you can’t directly reap these benefits.

Relatedly, utilities are now considering EVs as part of larger energy systems solutions, including connected to a smarter grid that would allow vehicle-to-grid (V2G) interactions as “mini-battery” storage facilities; good for backup power sources in emergencies, and for playing a role in ancillary markets. In other words, the value proposition of an EV is much more than switching out an engine block for a battery. And there are more benefits to having EVs on the road today than “just” their ability to get cleaner over time as the power sector decarbonizes.

The research itself at CityLab notes that full lifecycle analysis for EVs was not done apples-to-apples. So while it is a perfectly interesting and analytically helpful discussion, we should be careful when drawing conclusions. Or put differently, if you want our transport sector to be more sustainable, you need to consider three questions:

  1. What is the total mileage i.e. total number of cars * miles driven? This is the least related to the combustion-vs-EV debate, but is very important for the overall picture.

  2. What is the efficiency of the vehicle? In this case, EVs are about 60% more efficient. So irrespective of the power source, they’re still 60% more efficient.

  3. What is the CO2 per mile? For EVs, this of course depends on the power source and, as Holland’s research shows, today’s EVs will not be universally cleaner than gasoline or diesel cars. However, today’s EVs will support the switch to zero-carbon vehicles that we need in the longer term (as previously discussed).

If we look at any of these three questions in isolation, we’ll get a different answer to how we can best support the decarbonization of the transport sector. But by looking at all three together, we can consider the merits of different proposals appropriately.

In doing so, we acknowledge that EVs are part of the solution – not the silver bullet to building a sustainable transport sector – and they must be considered from a broader systems perspective, especially as they are directly connected to that system.

Tali Trigg is an energy analyst, technology policy advisor and writer. His work includes research and analysis on energy and transportation, with an emphasis on the role of cities in shaping transport energy demand and mobility solutions. His blog covers the wide range of mobility and energy, with deep-dives into numbers and maps, but is keen to cover anything transport-related and under-reported. Opinions are his own.

More by Tali Trigg