Electric Cars and Energy Independence, Part I

Will electric cars and tougher fuel economy standards put America on the super-fast track to oil independence? Not necessarily, says MIT in its recent report, On the Road in 2035 (pdf):

Transitioning from our current situation onto a path with declining fuel consumption and emissions, even in the developed world, will take several decades -- much longer than we hope or realize.

The delay, says researchers, is not for lack of technology but for the time involved -- the time to make a radical new technology ready for mass market, the time for such advances to become pervasive and the long wait for old cars hit the scrap heap.

In fact, even with aggressive market penetration rates of new technologies, it will be difficult to reduce the 2035 fleet fuel use by more than 10 percent below fuel use in 2000.

So, as strategies and policies abound for a more fuel-efficient US fleet, it’s time to assess their worth. Here are three of the latest:

1.) The CAFE standards of 2007, which legislate a 35 mpg standard for cars and light trucks by 2020.

2.) Barack Obama’s proposal to put one million PHEVs on the road by 2015.

3.) Former Intel chief Andy Grove’s plan to convert the existing light duty truck fleet to PHEVs.

Will they work and by when? The answers may surprise. First the MIT findings...

MIT Report in Brief

The MIT report was released in July 2008 by the MIT Laboratory for Energy and Environment as an update to their 2000 report, "On the Road in 2020." The authors considered a wide range of car and light duty truck changes that could be important in the next 25 years, the development time before they are ready for large-scale use and how quickly manufacturing can ramp up. Finally, they used a fleet model to determine the year-by-year impact as new vehicles replace old ones headed for the junk yard. They made reasonable assumptions for the time to do the R&D, work out the real world bugs, gain acceptance and build the factories to achieve widespread use.

The researchers considered better gas and diesel engines, gas hybrids, plug-in electric vehicles (PHEVs), battery electric vehicles (BEVs) and fuel cell cars. In addition to these kinds of dramatic technology changes, they also allowed for routine improvements in other areas like weight reduction and efficiency. They did a full lifecycle analysis of cars in 2035, considering not only the fuel they burned in their lifetimes, but also the energy used to produce the fuel, the energy used to manufacture the vehicles and the energy used when scrapping them.

They made several critical assumptions worth mentioning. The first is that gas would always be available at prices that wouldn't affect consumption. This is a business-as-usual approach that doesn't require US drivers to change their driving habits. We already know this assumption is wrong because as soon as gas prices went high enough, Americans started driving less, almost 5% less than last year for the month of June. For the purposes of the report, this is a reasonable assumption, and not necessarily because it's true but to establish a baseline. If gas prices are cheap enough, it lets us get a handle on what will happen. If gas prices remain high enough to discourage driving, we'll know to which consumption levels to start applying estimated changes.

Technology Development and Adoption

The MIT researchers found that even after a new technology is ready for widespread use, it doesn't become commonplace for quite some time. They analyzed historical technology deployment and found that it takes 10 to 30 years. For example, disk brakes took about 10 years. Fuel-injection stayed in the 5% to 10% range for about 10 years and only then started gaining market share, and in another 12 years, it approached 100% of new vehicles. If you owned a 70s era fuel-injected car, you know why it took so long -- it just didn't work that well. A more complicated change like variable engine timing took 30 years. The paper concludes:

Applying this logic to penetration of emerging propulsion systems across all market segments will yield at least a 15-20 year timeframe before they could garner a third of the market share.

Fleet Model

To estimate the impact of new technology on total gas consumption, you have to take into account all the existing vehicles. Even if all 16 to 19 million new cars sold next year in the US used no gasoline at all, we wouldn't stop using gas because of the old vehicles still being driven. Since new cars will be about 7% of the total vehicles on the road, even gas-free vehicles would reduce the total by no more than 7%.

The US light duty vehicles (LDV) fleet has 135 million cars and 105 million light trucks (including pick-ups, minivans and SUVs) for a total of 240 million. Once you consider the age of the old LDVs, the average mpg year-by-year, mileage patterns and retirement rates, you can build what they call a Fleet Model. The MIT researchers used historical data from 1960 onward to create the model.

Cars have been steadily improving in many ways since their quality nadir in the mid-1970s and are much more reliable and durable than before. The good news for owners is that lifetime has steadily grown to an average of about 17 years for cars and 16 years for light trucks. The bad news is that it will be about 17 years after any great new technology is universal before most of the old clunkers are off the road. The study:

There are a greater number of older vehicles on the road today, and they add to the inertia of the vehicle fleet.

And:

Fleet fuel use responds with a lag of some 10 years to changes in the new vehicle market.

In conclusion, this is a fleet that takes a long time to turn around, hence, even with pretty aggressive changes, the authors still expect gas consumption to keep rising until it peaks in 2020.

Favored Technology: PHEV

Of all the new technologies the study's authors consider, they are most optimistic about plug-in hybrids (PHEVs):

The plug-in hybrid offers important advantages over both fuel cell and battery-electric vehicles with respect to fueling infrastructure, vehicle range, and technological risk. First, it does not require changes to the fueling infrastructure on the same scope as either the fuel cell, which would require extensive ramp-up in hydrogen production and distribution, or as the electric vehicle, which would likely require rapid-recharge electric fueling stations and major upgrades to the electricity generation and distribution infrastructure. Second, it is not range-limited in the same sense as an electric vehicle.

They expect PHEVs to run on gas about half the time and use the rechargeable battery the other half.

PHEV simulations are based on a vehicle with a 30-mile (50 km) electric range, which is estimated to capture approximately 50% of the vehicle's total miles.

Will the electric power plants be able to recharge this many cars? I've seen a wide range of estimates on that subject, but here's their take:

In the Hybrid Strong Scenario, the market share of plug-in hybrids grows to 15% of new LDVs in 2035. The total electricity demand by plug-in hybrids in 2035 grows to 59 billion kilowatt hours. As the fleet of PHEVs grows, the demand for electricity will increase by approximately 6-10 billion kilowatt hours in the decade after that. The current electricity consumption in the United States is approximately 3,700 billion kilowatt hours, and is projected to increase to over 5,200 billion kilowatt hours by 2035 [EIA 2007a]. Therefore, plug-in hybrids will represent only 1-2% of electricity demand under this scenario...

In spite of the big thumbs-up, the authors still estimate that it will take 30 to 35 years before the full potential of PHEV is achieved. They expect 8 to10 years for the technology to mature, 15 years for widespread adoption and another 15 years for fleet turnover.

In contrast, they don't expect much from pure electric or fuel cell cars:

At the same time, other advanced technology vehicles, including hydrogen fuel cell or battery electric vehicles, will continue to suffer from high cost and other limitations. Their limited market penetration means that their impact on fuel use and emissions is unlikely to be significant over the next few decades.

The authors expect that battery electric vehicles (BEVs) will remain too expensive and unpopular because of another important assumption that might just be wrong -- that Americans will demand at least a 200-mile range. Battery weight and cost go up rapidly with range, so while a 200-mile BEV might be too expensive, a 50-mile range car would be much more reasonable. The issue is one of marketing and not technology: Will US drivers buy a car with limited range but one that is still adequate for 90% of their trips? The study authors think the answer is no.

Fuel Alternatives

To be able to calculate the impact on greenhouse gas emissions, they also had to consider how the gas would be produced and refined and to look at alternative fuels. No good news here either:

Alternative fuels that replace petroleum fuels are unlikely to change GHG emissions significantly. The major near-term alternatives are based on fossil raw materials like the Canadian oil sands or coal, which increase GHG emissions. Some biofuels may prove beneficial, depending on the particular biomass feedstock and conversion technology. But the U.S. emphasis on corn-based ethanol is not obviously justifiable. It has high economic costs, questionable GHG advantages, and other unfavorable environmental impacts.

They do say that cellulosic ethanol might help, but we'll have to wait and see how that develops.

Recommendations

The researchers offer a few recommendations. Aside from technical advances, the key to reducing gas consumption is to use the advances to improve gas mileage instead of making light cars and trucks bigger or faster, as has been the norm:

In the United States, the emphasis on enhanced performance-and to a lesser extent, increases in vehicle size have been so strong that no significant fuel consumption gains have been realized over the past 25 years. In Europe, the emphasis on performance has not been as strong, and some half of the fuel consumption improvements that could have been realized have already been achieved. Using half of all future efficiency improvements to reduce fuel consumption rather than emphasizing performance would alone reduce fuel use by 13% in 2035. Using all future efficiency improvements to lower fuel consumption would reduce fuel use by 26% in 2035.

Their second recommendation is to get to work as soon as possible. It is discouraging that it will take so long for improvements to percolate through the fleet, but that makes it even more important to get to work now.

CAFE Standards: 35 MPG by 2020

The Energy Independence and Security Act of 2007 increased the CAFE (Corporate Average Fuel Economy) standards to 35 mpg by 2020 for cars and light trucks. Although the Act set the eventual CAFE level to 35 mpg, it didn't legislate a set of intermediate standards to get to that level, a detail that was left to the DOT to figure out. Since we don't have a precise schedule of changes, we can't say for sure what the exact impact will be, but we can guess that the standard will go up about 4% per year and come up with a ballpark result.

It's not hard to calculate the mpg of the old LDVs that will be retired in the next 12 years because, as the MIT study points out, gas mileage has stagnated for the last 20 years. They get about 20 real world mpg.

Of course, everyone knows "your mileage will vary" from the EPA standards, and for years everyone has also known that the peculiar EPA meaning of "vary" is that you won't come close to that big number on the new car sticker. Estimates of how much your mileage will "vary" usually run 10% to 20%, so let's discount the 35 mpg target by 15%, or 5 mpg, to account for the variance. If today's new vehicles average 20 mpg and in 2020 actually get 30 mpg, the average new member of our auto fleet over the 2011 to 2020 model years will get about 25 mpg. Replace 180 million LDVs getting 20 MPG with 180 million getting an average of 25 MPG, and in 2020 gas consumption should be down about 15% from today's levels.

This assumes that the miles driven do not increase, which has been a pretty bad bet for the last 50 years. Ignoring 2008 for the moment, in the last few years, the annual increases in miles driven have been about 1 percent. If it continues at that rate, actual gas consumption in 2020 might only drop by 4% from today's level. As I pointed out earlier, miles driven have gone down due to high gas prices, and in fact, the DOT says they've been dropping since last November. Miles driven in June were nearly 5% less than in June 2007. If gas prices stay high, the fewer miles driven may continue, but the idea of what price makes gas too expensive tends to be elastic, and if Americans get used to $4.00 a gallon gasoline, they will probably start driving again.

To conclude, the fleet turnover delay requires us to look past 2020 to see the full impact of the CAFE changes. Allow 17 years for all the pre-2020 LDVs to be junked. Then, based only on these CAFE changes and holding miles driven at 2007 rates, in 2037 we would expect the entire fleet to be getting 30 MPG (the nominal 35 MPG standard minus the variance), and gas consumption will be about 33% lower than 2007.

What can we expect from the plans of Barack Obama and Andy Grove? That’s the question I’ll explore in "Electric Cars and Energy Independence, Part II."

• ‹ previous
• up
• next ›