On June 22nd Scientific American rolled-out a Web-only article titled “The Dirty Truth about Plug-in Hybrids, Made Interactive” that summarizes a January 2008 report from Oak Ridge National Laboratory and shows why plug-in vehicles in the U.S. will, on average, be just a little bit dirtier than gasoline HEVs.
You read that right – dirtier, not cleaner!
I first raised the issue in an August 2009 article titled PHEVs and EVs, Plugging Into a Lump of Coal, where I estimated that plug-in vehicles would be about 25% cleaner than HEVs, but the marginal cost of CO2 abatement with plug-in vehicles would be five times higher than the marginal abatement cost with HEVs. The Oak Ridge report went a couple of levels deeper than my simple calculations and evaluated:
- Baseload power requirements and generating facilities in 13 regions in the year 2020;
- The specific types of generating facilities that would be used to charge plug-in vehicles; and
- The regional CO2 increase or decrease from using those generating facilities to charge plug-ins.
The following graph highlights the comparative CO2 increase or decrease in the 13 regions identified in the Oak Ridge study and discussed in the Scientific American article.
After accounting for the projected number of vehicles in each region, the national average was a 0.37% increase in CO2 emissions. Given the modest CO2 reductions from plug-in vehicles in regions like the Pacific Northwest and the significant CO2 increases in the industrial heartland, it would be easier, cheaper and better policy to use domestic natural gas in HEVs and forget about plugs entirely. Where HEVs cut to the heart of the CO2 problem nationwide, plug-ins only nibble around the edges in a few select regions.
Last month the American Chemical Society published a similar white paper from Tsinghua University, Beijing, and the Argonne National Laboratory Center for Transportation Research titled “Environmental Implication of Electric Vehicles in China,” which concluded that implementing electric vehicles in China would increase CO2, SO2 and NOX emissions. It also concluded that gasoline HEVs are more environmentally friendly, more commercially mature and less cost-intensive. The following graph comes from page 4 of the white paper.
While the CO2 emissions data from both China and the U.S. is damning, simple calculations prove that electric vehicles like the Leaf from Nissan (NSANY.PK) and the MiEV from Mitsubishi (MMTOF.PK) save an average of 10.4 gallons of gasoline per year for each kWh of incremental battery capacity while PHEVs like the Volt from General Motors save an average of 7.6 gallons of gasoline per year for each kWh of incremental battery capacity.
I’ll encourage each of you to run your own discounted cash flow calculations on annual gasoline savings of 10.4 and 7.6 gallons per kWh and then compare your calculated value with current battery costs of ~$1,000 per kWh and projected future costs of ~$500 per kWh. I’ve run the numbers and am not impressed.
In addition to Tesla Motors, which is scheduled to go public next week at a price of $14 to $16 per share, there are six pure-play battery companies that focus on electrification solutions for transportation. A123 Systems (AONE), Ener1 (HEV) and Valence Technologies (VLNC) are all working on lithium-ion battery solutions for plug-in vehicles. Maxwell Technologies (MXWL) is working on supercapacitor solutions for city bus, stop-start vehicle and HEV applications. Exide Technologies (XIDE) and Axion Power International (AXPW.OB) are working on advanced lead-acid and lead-carbon battery solutions for stop-start vehicle and HEV applications.
The following table assumes the Tesla IPO will go off at $15, the mid-point of the price range, and includes summary balance sheet and market valuation metrics for all seven companies. For both working capital and stockholders equity, the table reflects the dollar amount and the percentage of market capitalization those values represent. The Blue Sky column highlights the spread between market capitalization and stockholders equity.
It doesn’t take much training to see that valuation premiums are much higher for plug-in vehicle companies than for lead-acid companies. In March of this year I suggested that stop-start idle elimination and other vehicle efficiency technologies were the investment equivalent of birds in the hand while plug-in vehicles were a flock of wild geese on the wing. In April I reported that the EPA and NHTSA were forecasting a 42% market penetration for stop-start systems by 2016. Over the last month we’ve seen important articles in prestigious publications expose the zero-emission myth as urban legend. With Oak Ridge and Argonne Laboratories, the American Chemical Society and Scientific American targeting the wild geese with double-barreled shotguns, I’m more convinced than ever that the market will soon shift to a more rational focus on economic reality and business opportunity.
Disclosure: Author is a former director of Axion Power International (AXPW.OB) and has a substantial long position in its stock.
A couple factors I don’t think this study considered:
1. PHEVs are likely to be smaller, lighter, and aerodynamic than HEVs, because the cost of batteries means that the se efficiency measures are necessary to achieve decent range at a (relatively) affordable cost.
2. The electricity mix used for calculating EV CO2 footprint should be incremental generation, not current mix. What type of new generation will be added to meet the new demand from HEVs? Given current trends, the bulk of it will be wind, gas, and Demand Response, all of which are much cleaner than the existing electricity mix.
The ORNL study that was the basis for the Scientific American report is incredibly complex and detailed. I think you’re probably right about them not taking future configuration differences into account. From everything I’ve read, it looks like they did focus on the incremental generating assets that would most likely be used to satisfy charging demand, but expectations about what new generating assets will be a decade from now are always pretty murky, so if the actual additions are different the result may also be different.