In June 2012, the UCLA Institute of the Environment and Sustainability, working under contract for the California Air Resources Board, published a fascinating "life cycle Analysis Comparison of a Battery Electric Vehicle and a Conventional Gasoline Vehicle" that proves far more than the UCLA researchers or CARB bargained for.
It proves that short-range BEVs can be marginally better for the environment if you're willing to ignore end-user economics, but long-range BEVs like the Model S from Tesla Motors (NASDAQ:TSLA) are an economic, energy and emissions abomination.
The UCLA study included detailed life cycle energy input and CO2 emissions assessments for two vehicles. It used a $15,000 Nissan Versa as a baseline conventional vehicle, or CV, and a $35,000 Nissan Leaf as a comparable battery electric vehicle, or BEV. It then reported life cycle energy input and CO2 emissions data, together with relevant weight information, in each of the following categories:
- Vehicle parts;
- Battery and Engine;
- 180,000 miles of driving; and
While the study also included a limited data set for a hybrid electric vehicle, or HEV, the data set was based on "weighted averages and extrapolation" instead of detailed analysis. So I'll ignore the HEV estimates for purposes of this article.
Since I'm using the UCLA study as the scholarly authority for this article, I have not parsed or modified the data. I will note, however, that the UCLA study incorporates several assumptions that tend to cast BEVs in an unduly favorable light including:
- Using California's grid mix as the benchmark for electricity inputs despite the fact that the US grid mix is 29% more energy intensive and 61% more emissions intensive;
- Assuming that only 50% of the battery will need to be replaced over 15 years and 180,000 miles of daily driving; and
- Assuming comparable recycling energy inputs for lead-acid and lithium-ion batteries
The two most important graphs in the UCLA study were the "Energy Inputs life cycle Comparison" on page seven and the "CO2 Equivalents life cycle Comparison" on page eight. In both graphs, the left-hand column is the CV, the center column is the BEV and the right-hand column is the HEV.
The energy inputs graph shows that the CV consumed about 850,000 Megajoules of energy over its useful life while the BEV only consumed 500,000 Megajoules. Similarly, the CO2 emissions graph shows that the CV will generate about 63 metric tons of CO2 over its useful life while the BEV will only generate 32 metric tons. If energy inputs and CO2 emissions are the only things that matter, the BEV is a marginally better choice on both counts. When it comes to economics, however, UCLA found that it would take 13 years for the purchaser of BEV to recover his cost premium and break even. On a discounted present value basis, the study concluded that a BEV would be roughly 7.5% more costly than a CV over a 15-year useful life.
If you study the UCLA graphs for a minute, you'll see that the second largest contributor to BEV energy inputs and CO2 emissions is the lithium-ion battery pack that a BEV uses as a fuel tank replacement. In a short range BEV like the Leaf, the energy inputs and associated CO2 emissions look pretty reasonable to a casual reader. The question UCLA never asked is, "What happens to the relatively benign energy inputs and CO2 emissions figures if you use a larger battery pack to eliminate range anxiety?"
In a perfect world UCLA would have presented tabular data in addition to their energy inputs and CO2 emissions graphs. Since UCLA didn't present the tabular data, I downloaded a graph digitization application from GraphClick and used that software program to break down the stacked column data for CVs and BEVs into the five categories used in the UCLA study. Since graph digitization depends to a degree on the visual acuity of the user, I ran five different sets of digitization records and used the average of the five sets for this article. For readers who want to review my raw digitized data I've posted a downloadable Excel workbook to my Dropbox.
After carefully considering the UCLA data categories and study methodology, I believe it's reasonable to assume that:
- Non-drivetrain vehicle parts and transportation are stable values that won't change as you increase battery size to offer a longer range;
- Energy inputs and CO2 emissions for the battery pack will increase proportionally as you increase battery size to offer a longer range; and
- Values for the use and disposal phases will increase proportionally with vehicle mass.
Based on those assumptions I conducted a sensitivity analysis based on the UCLA data that compares a CV against their base case BEV-73 and three hypothetical cases for a BEV-146, a BEV-219 and a BEV-291. The calculations for my graphs are included in the Excel workbook.
My first graph uses the UCLA data to show the comparable life cycle energy inputs for a CV in the left-hand column followed by four BEVs with successively larger battery packs.
My second graph uses the UCLA data to show the comparable life cycle CO2 emissions for a CV in the left-hand column followed by four BEVs with successively larger battery packs.
In both cases, the first battery pack increment nullifies the substantial bulk of the BEV's energy and emissions advantages while making the economics worse. By the time you get to a 200-mile BEV, the allegedly green solution is less than 10% cleaner than a comparable CV. If you push the BEV's range to almost 300 miles, it's 10% more energy intensive and 20% more emissions intensive.
As the final step in my analysis I did a rough calculation of the CO2 breakeven points, the moment in time when the reduced use phase emissions of a BEV fully amortized the higher manufacturing phase emissions. For UCLA's basic BEV-73, the crossover point arrives late in the third year of driving. For a hypothetical BEV-146 the crossover point arrives in the middle of the eighth year. The BEV-219 reaches carbon breakeven in the eleventh year of ownership and the BEV-292 never amortizes its carbon debt.
There are many times in life when the failure to ask the right questions leads to a remarkably wrong conclusion. One of the most glaring examples I can imagine is the CARB's ZEV credit regime that allocates 2 ZEV credits to BEV manufacturers like Nissan (OTCPK:NSANF) who build marginally cleaner vehicles and allocates up to 7 ZEV credits to BEV Manufacturers like Tesla who build economic, energy and environmental abominations like the Model S. While CARB has recently deferred revisions to its ZEV credit regime that would change the treatment of quick-change battery swapping schemes, what it should do is revisit the fundamental credit structure that gave Tesla an unmerited $158 windfall over the last 12 months for manufacturing long-range BEVs that are far dirtier and far less energy efficient than their conventional counterparts.
I love a good story as much as the next guy and there are few stories in the market that are more appealing than Tesla's. Seriously, how could anyone criticize a manufacturer of electric cars that offer market-leading range on a single charge combined with attractive styling, neck-snapping acceleration and zero tailpipe emissions? The answer is simple; the zero tailpipe emissions part of the story doesn't even rise to the level of a half-truth because the manufacturing phase energy inputs and emissions negate any possible benefit on a life cycle basis.
Tesla manufactures a fine performance car that is no greener and no cleaner than any other performance car. The Model S and the upcoming Model X are economic, energy and emissions abominations. The promised third generation Tesla with a 200-mile range will be significantly dirtier than its conventional counterparts if it's ever built. When the truth about Tesla's energy inefficiency and its dreadful life cycle emissions profile becomes more widely understood its stock price will undoubtedly plummet into the $20 range, which is the only reasonable price range for a money losing niche manufacturer with a book value of $4.60 per share.