Comparing Driving Costs of EVs and Conventional Cars

• A new Department of Energy website helps consumers compare the energy costs of EVs to non-plug-in cars by converting kilowatt-hours into "eGallons".
• How valid this proxy is depends heavily on assumptions about the cars being compared to EVs.  If hybrids set the bar, then DOE's eGallon prices are significantly understated.

I’ve been looking through a new website developed by the US Department of Energy (DOE) to assist consumers in comparing the energy costs of driving an electric vehicle (EV), relative to posted gasoline prices in their state. I heard about this site at the US Energy Information Administration’s (EIA) annual energy conference in Washington, DC last month. It sounded like a handy tool for both current EV owners and those considering buying one, but I couldn’t help thinking about it in the context of a presentation I saw at the same conference on the cost effectiveness of federal tax credits for EV purchases. A key question in both instances concerns just what kind of car is being replaced by that new EV.

The website uses simple math, together with the EIA’s continuously updated data on gasoline and electricity prices around the country, to come up with a national and state-by-state price for an “eGallon”. This imaginary construct is essentially the quantity of electricity that would take a typical EV as far as a gallon of gasoline would take the average new conventional car. As the text points out, it’s hard for consumers to calculate this for themselves. They see gasoline prices everywhere they drive but must dig through their utility bills to find their electricity price–not always obvious–and then might not know how to compare the two.

The site’s documentation indicates the eGallon calculation is based on the average energy usage of five specific EVs, including the Chevrolet Volt, Nissan Leaf, and Ford Focus EV, along with the 2012 EPA fleet average fuel economy for what EPA defines as small and mid-size cars. The result is side-by-side postings of the US average gasoline and eGallon prices, plus a drop-down menu to replicate that for each state. The site also includes the chart below, comparing these two prices over the last decade.

Two facts become immediately apparent. First, electricity is generally a cheaper fuel for cars than retail gasoline. That’s true for a variety of reasons, including the higher end-use efficiency of electric motors compared to internal combustion engines and the lower cost of most of the fuels used to generate electricity in the US. For example, the natural gas burned in power plants sold for the equivalent of $ 20.40 per barrel last year, while the global benchmark for oil averaged nearly $112/bbl. It also appears to be less volatile, at least at the level of national averages.

However, just as there’s no single gasoline price for the whole country, neither is there a single electricity price. Even the state averages used by the DOE to calculate eGallon prices mask a bewildering variety of regional electricity price tariffs and tiers. So your cost to recharge an EV might not just vary by location, but by time of year, time of day, and the specific rate plan that applies to you.

My main concern about the site derives from something much simper: the big central assumption that EVs compete with the average cars sold in America last year. According to the eGallon site, the average small-to-medium US car in 2012 got 28.2 miles per gallon (mpg) in combined city and highway driving. Using that figure, and with residential US electricity prices averaging 11.6 ¢/kilowatt-hour (kWh) in March 2013, the national eGallon price for March would have been $1.14/gal., compared to $3.71/gal. for unleaded regular gasoline. But what if we assumed that the cars most often compared to a new EV were not average cars, but other efficient cars, as logic and my intuition suggest? If we substituted the fuel economy data for a conventional Ford Focus or Toyota Prius hybrid, the eGallon price would jump to $1.26 or $2.03, respectively.

In some respects this result is fairly obvious. If you were already contemplating buying a hybrid, an EV won’t save you as much as if you were thinking of buying a conventional mid-size sedan. However, this distinction is important enough that the DOE should consider refining its eGallon calculator. EVs are much like wind and solar installations that cost more than conventional alternatives, but are expected to produce over their lifetimes economic or environmental benefits that offset those higher costs. The attractiveness of that big up-front investment is directly proportional to those benefits. I don’t have the data that would clarify the actual comparisons EV buyers are making, but someone must, perhaps including DOE. And it turns out that this isn’t just important for calculations like eGallon, but also for assessing the cost-effectiveness of federal EV policy.

That brings me to the Congressional Budget Office’s analysis of federal EV tax credits last fall. The report merits a posting of its own, but one nugget I gleaned from the presentation at the EIA Conference was that the CBO found that the current federal credit of up to $7,500 per car was still insufficient to make most EVs cost-competitive on a full-life basis with conventional cars. Yet despite this, the effective cost to taxpayers of each gallon of gasoline saved by a Leaf-type EV was well over $6 when compared to conventional cars getting average fuel economy, and over $10 vs. high fuel-economy compact cars. That’s assuming they save any gas at all, because of the way the Corporate Average Fuel Economy rules have been structured. Implied costs for greenhouse gas emissions avoidance were even more startling, at over $400/ton of CO2 in most cases.

The desirability of a tool like “eGallon” is rooted in the convoluted way we talk about transportation fuel economy and energy costs in this country. Miles per gallon is itself a poor metric, compared to something like gallons per 100 miles, or even miles per dollar. That's because it obscures the high value of modest improvements in high-consumption vehicles, while exaggerating the value of shifting from very efficient to ultra-efficient cars. It’s also more useful for policy makers than consumers, who are ultimately concerned about outcomes in dollars per mile or dollars per trip.

Recognizing the impracticality of training 300 million consumers to think about this subject differently, eGallon might prove useful, but only as long as it is grounded in the best information we have about the vehicle choices that potential EV buyers are actually considering. Since current EV incentives apparently provide a poor return to taxpayers, an overly simplistic tool that drives consumers too far in that direction might be worse than not having such a tool at all.

A different version of this posting was previously published on Energy Trends Insider.