GM restructures global vehicle line teams; former electric car director heading Product Programs
Ford prices Lincoln MKZ below new Lexus ES 350, with no premium for hybrid model

Researchers suggest subsidies and policies targeting plug-ins with small battery packs would produce more benefits at lower cost

Federal subsidies and policies to encourage plug-in vehicle adoption would produce more benefits at lower cost by targeting the purchase of vehicles with small battery packs, according to Jeremy J. Michalek, an associate professor at Carnegie Mellon University and his colleagues Mikhail Chester, an assistant professor at Arizona State University; and Constantine Samaras, an engineer at the RAND Corporation.

Their paper, to be published in the summer issue of the journal Issues in Science and Technology, is based on their earlier study that found that strategies to promote adoption of hybrid electric vehicles (HEVs) and plug-in hybrid electric vehicles (PHEVs) with small battery packs offer more social benefits (i.e., air emissions and oil displacement benefits) in the near term per dollar spent than PHEVs and battery-electric vehicles (BEVs) with large battery packs providing longer electric range. (Earlier post.)

The electrification of passenger vehicles has the potential to address three of the most critical challenges of our time: Plug-in vehicles may produce fewer greenhouse gas emissions when powered by electricity instead of gasoline, depending on the electricity source; reduce and displace tailpipe emissions, which affect people and the environment; and reduce gasoline consumption, helping to diminish dependence on imported oil and diversify transportation energy sources.

...existing and proposed subsidies provide larger payments for vehicles with larger battery packs. Larger battery packs enable vehicles to displace more gasoline, so at first glance one might think that subsidizing larger battery packs is better for the environment and for oil security. But large battery packs are also expensive; the added weight reduces efficiency; they are underused when the battery capacity is larger than needed for a typical trip; they have greater charging infrastructure requirements; and they produce more emissions during manufacturing. Whether larger battery packs offer more benefits on balance depends on their net impacts from cradle to grave.

—Michalek et al. (2012)

In their earlier study, published in the Proceedings of the National Academy of Sciences (PNAS), Michalek and his colleagues quantified lifetime externality costs, including greenhouse gases, human health effects, agricultural losses, and infrastructure degradation, caused by air emissions from conventional and electrified vehicles.

They found thousands of dollars of damages per vehicle (gasoline or electric) paid by the overall population rather than only by the emitters. They also found that plug-in vehicles could only modestly reduce those costs—an amount much lower than the $7,500 federal tax credit and small compared to ownership costs. This, they explained, is because the damages caused over the lifecycle of a vehicle are caused not only by gasoline consumption, but also by emissions from battery and electricity production, which are increased with plug-in vehicles.

Current federal policy is weighted toward plug-in with larger battery packs. The Chevrolet Volt with a 16 kWh pack and ~35 mile electric range receives a $7,500 credit. The Toyota Prius Plug-in Hybrid with a 4.4 kWh pack and ~11 miles electric range receives $2,500.

At first glance, tripling the subsidy may seem justified because the electric range is tripled. But tripling the range does not mean tripling the amount of gasoline displaced or emissions reduced: Increasing battery size has diminishing returns. In fact, when we consider US driving patterns (many short trips, where the larger battery is only dead weight), US average emissions from battery and electricity production, and the other factors described above, the small 4.4-kWh battery actually has more net benefits than the larger 16-kWh battery.

Even in the most optimistic scenarios where vehicles are charged with zero-emission electricity, the larger battery packs offer only comparable or slightly greater net benefits, not double or triple. Public funds are limited, and because today’s policy consumes more resources when subsidizing large-battery vehicles, fewer of them can be supported under a fixed budget. Allocating a fixed budget to a flat $2,500 subsidy for all plug-in vehicles would more than triple the potential air-emissions and oil-displacement benefits of the subsidized vehicles as compared to subsidizing one-third as many large-battery vehicles at $7,500 each.

—Michalek et al.

Advances in batteries may allow future large battery packs might be able to offer the largest benefits at the lowest costs, the note, adding that policies supporting R&D for battery improvements and large emissions reductions from electricity generation can help move the country in this direction. This may take decades to realize, however, and is uncertain, they argue.

In the near term, they suggest, HEVs and PHEVs with small battery packs are more robust, offering more air-emission and oil-displacement benefits per dollar spent.

We should not forget that the most efficient policies would target externalities directly, through mechanisms such as an economy-wide carbon price, cap-and-trade policies, and gasoline taxes. Such policies are generally understood to be far more efficient than technology-specific subsidies, and we should consider subsidies as an inferior substitute given the political difficulties of implementing efficient market-based policies that address the problem directly. In the absence of such policies, federal subsidies and policies designed to encourage electrified vehicle adoption would produce more benefit at lower cost for the foreseeable future by targeting the purchase of vehicles with small battery packs.

—Michalek et al.

Issues in Science & Technology is a forum for discussion of public policy related to science, engineering, and medicine. Issues is a publication of the National Academy of Sciences, the National Academy of Engineering, the Institute of Medicine and the University of Texas at Dallas.


  • Jeremy J. Michalek, Mikhail Chester, Constantine Samaras (2012) Getting the Most Out of Electric Vehicle Subsidies. Issues in Science and Technology

  • Michalek, J.J., M. Chester, P. Jaramillo, C. Samaras, C.-S. N. Shiau, and L. Lave (2011) “Valuation of life cycle air emissions and oil displacement benefits of plug-in vehicles,” Proceedings of the National Academy of Sciences of the United States of America, DOI: 10.1073/pnas.1104473108



Talk about people who want to shoot themselves in the foot ...


This makes a lot of sense.
Batteries are expensive and heavy.
The first few KwH save the most fuel.

You will save more fuel with 4 4KwH PHEVs than one 16 KwH PHEV, so the subsidies should reflect this.

If batteries get cheaper, maybe you could review the subsidies, or maybe just abolish them (assuming PHEVs take off).


Absolutely. Look at the VW Jetta Hybrid. 30% mileage improvement with 1 KWH. It's in everybody's interest to land that 30% improvement, and get the scale manufacturing benefit of all the other hybrid components, in as many vehicles as possible.


The problem of small battery pack is that they are not suited to accelerate the car or sustain highway speed, they just don't provide enough power for all electric operation. No battery can both give high power en high energy density today, it is not only a problem of limited range

Roger Pham

The Ford C-max hybrid seems to be in good balance.


Many of the author's observations are correct but his conclusions are flawed. First, he talks about "near term" which is useless if he wants to influence future policy decisions (which would be outdated if based on today's world). Second, PHEVs are inherently inefficient and therefore must be seen as a transitional technology; there are two complete propulsion drivetrains--while one operates, the other is dead weight. If you lost the ICE in a PHEV, that weight would be gone and allow the weight of the vehicle structure to be reduced proportionally. With the same battery, the car would have greater range and performance, and would satisfy most all trip requirements. This resulting short-range (say, 50-60 miles) BEV would not be for everyone, certainly, but would work well for many.

I'm not suggesting that all BEVs have 50-60 mile range; 100-mile range BEVs are good as well. There should be market choice--a range of ranges.

All the talk of range anxiety and the need for long-range highway BEVs is misplaced and unhelpful. We've got ICE vehicles and HEVs for people who need to travel long distances frequently. Modern BEV technology is in its infancy -- little steps first.

Incentives for BEVs promote the development of the technology for the future; that, I thought, was their purpose.


This is exactly what Toyota has been saying and doing.
Toyota people are very smart?

A good solution would be to use modular battery packs, start with a small 4 Khw module and add more modules latter, if really required and/or when batteries price have dropped.


The study is incomplete. The larger batery the lesser battery cost per kWh. The other issue is that the smaller battery the bigger emphesis on ICE and ICE drivertrain complexity required. After certain capacity of battery achieved the EREV engine could be just marginal and just 'in case' therefore reducing drivertrain cost.
The second issue - technological subsidies and governmental involvment are needed. Private business is not capable of making any major shifts. In Europe we have high fuel taxes but no major drivertrain advances in place.

Roger Pham

A big battery pack is also dead weight. A PHEV allows very long range and quick fill-ups while allowing space saving and weight saving and cost saving by the use of a smaller battery pack. The smaller battery pack of the C-max Energi allows more internal space and 5-seat vehicle in comparison to the Volt that has twice the battery pack size but less cargo space and can only seat 4, while is heavier.


The idea is not to obtain the biggest reduction possible in oil use by funding the largest number of vehicles, but to provide the maximum stimulus to technologies which can in future greatly reduce it long after the subsidies have ceased.

If you ask the wrong question, you get the wrong answer, which is what has happened to the people who did this study.


But Roger, the ICE and all it's accompanying parts cost about $3,000 in a modern low or mid end car, they take up lots of space (engine, transmission, gas tank, larger radiator, etc etc)and add a lot of weight as well. They don't add as much weight as additional batteries but they do offset it by some amount (between 300-500lbs depending on the car) for the whole ICE get up.


I do understand all their reasoning on this, but if they really want to make a dent in gas consumption, then require that all cars sold in America have stop/start capability. That would have the most return on investment if you're trying to reduce oil consumption. I'm a huge EV nut, but I'm willing to admit that.

And change the stupid EPA ratings to reflect this! Right now, they've set the test up so that a manufacturer gets ZERO credit for this despite the fact that in the real world it makes a big difference in normal driving.


One thing that hinders the most wider adoption of PHEVs with small battery pack is the lack of high-energy (20+ kWh/kg), low price (below $2,000/kWh) ultracapacitors. I think currently prices are $15,000/kWh, and densities 5 kWh/kg.
Treehugger IMHO identified the real problem - small battery pack doesn't have enough energy for acceleration. Current batteries are either high energy/low power or high-power/low-energy. It is very likely to remain that way in the future, although both energy and power densities will go up.
Toyota HSD goes around it by using relatively strong generator (MG1), ICE driven, when accelerating, as small battery cannot provide enough power. Once ultracapacitor prices come to levels mentioned above, HSD system will probably become obsolete, large MG1 will be a wasted resource.
It is proven that the optimal HEV design would use a COMBINATION of low-energy and high-power ultracapacitor (200-400 Wh energy, for passenger cars, over 200,000 recharge cycles) and a battery that can provide about 1/3 of max e-motor power (or more).
Ultracapacitor module and battery pack are connected with two-way DC-DC converter, sized for max battery module power. Motor inverter (two-way) is connected to the ultracapacitor module.
For non-plug-in HEVs, no batteries will be used, just ultracapacitors that can last 10-15 years easily. So people will be buying HEVs without fear that several years down the road, they'll face an expensive battery replacement bill (now over $ 4,000).


Correction - relates to energy density stated in the first few lines of the comment (Wh/kg replaces kWh/kg):

One thing that hinders the most wider adoption of PHEVs with small battery pack is the lack of high-energy (20+ Wh/kg), low price (below $2,000/kWh) ultracapacitors. I think currently prices are $15,000/kWh, and densities 5 Wh/kg.


Another factor nobody's mentioned is that up to a point, the larger the battery pack the more effective regenerative braking gets.  So long as batteries remain cheaper than ultracaps, that's going to set a floor on the effective size of hybrid batteries.

Roger Pham

@MG, Treehugger,
The following is a link to a hobby-grade LiFePO4 battery capable of 30C continous discharge and 40C burst discharge for 10 seconds. The cost is ~$600/kWh retail at small size, but perhaps below $500/kWh in larger primatic size for automotive use. The specific energy density is 113 Wh/kg and the power density is 4.5 kW/kg. Battery of this type are used and abused extensively by R/C car racers so you can bet on the fact that the specs are correct and the battery can deliver and is durable, or it won't sell!

You can look at this link and do your own calculation.

What this means is that the Ford C-max Energi with 8 kWh pack on board, if equipped with this pack, will be capable of 8 x 40C = 320 kW of power in 10 sec burst, or 240 kW of continous power!!! Wow
The pack will weigh about 70 kg and will cost about $4,000. Very reasonable numbers. LiFePO4 can be charged 2000-3000 times. If you would further do the math, you will see that the cost of battery energy delivered to the drive shaft will be 1/2 that of petrol ICEV per kWh.

A HEV with a 1.3 kWh pack of this LiFePO4 powerhouse can expect burst power of 1.3 x 40C = 52 kW!!! Quite enough to get the car going without resorting to the ICE, for most of normal daily driving, unless you're doing traffic-light drag racing.


What davemart said:

"he idea is not to obtain the biggest reduction possible in oil use by funding the largest number of vehicles, but to provide the maximum stimulus to technologies which can in future greatly reduce it long after the subsidies have ceased.

If you ask the wrong question, you get the wrong answer, which is what has happened to the people who did this study."


Herm, Davemart: the paper mentions minimizing cost not maximizing number of affected vehicles - if you start with the wrong constraint then you end up with the wrong question. A large number of technological challenges involved in progressing to PHEV are common to a HEV, such as electrification of non-drive systems and even some of the drive systems. So stimulating HEV in the short term is a reasonable strategy even if the longer term goal is the adoption of PHEV.


RP.....the battery you mentioned already exist and would do a very good job...imaging what improved future batteries with 3x to 5x the performance will do!

An ultra-cap/battery combo could improve performances today but may not be required in 5 to 10 years down the road. Higher performance batteries will give improved acceleration, better braking energy recovery, longer life and lower cost.

Very interesting decade ahead for interim HEVs/PHEVs and specially for BEVs.


Now that's what I'm talkin about...

The smaller battery pack PHEV is ideal; serves more households with a lifesaving backup power supply and smarter grid; increases household energy conservation; better match with rooftop photovoltiac; economic incentive to drive less & walk, bike, use transit more to patronize, diversify and strengthen local economies.

Ultra-capacitors do not offer these advantages. Both motorized vehicle and passenger train propulsion gain more advantages with HYBRID-drive rather than All-electric. Zero emission purists miss the point. The internal combusion engine is most advanced in a hybrid drivetrain.


You could offer a sliding scale of tax credits based on the mpg / electric range of a vehicle to offset a slightly higher price. Maybe $1k per kWh of battery capacity up to a max of $5k / 5kWh.

With regards to a small battery not being enough to accelerate the vehicle, I don't think it matters as you are carrying an engine with you which can provide acceleration and cruise while saving the battery capacity for 'city' situations

Chad Snyder

I have to laugh when I read things like "If you ask the wrong question, you get the wrong answer, which is what has happened to the people who did this study."

No offense, Davemart, because I've read many of your comments through the years and I know you're a very smart person.

But what is the question?

In my opinion, it should have something to do with real change.

Today, there are hundreds of millions of vehicles in the US fleet. Thus, how many years does it take to replace the current fleet?

Furthermore, in how many years will big battery plug-ins dominate even just yearly sales?

According to the DoE, a 100 mile range EV might be cost-effective with a gas equivalent vehicle without any tax credits by around 2025. That's a long time from now and, potentially, a lot of big tax credits.

Even if all cars are large battery plug-ins in 2025 -- whether EV or REEVs -- how long to replace just half the entire US fleet?

At some point the question has to include something about the legacy effect of the current fleet and how long it takes to replace.

In that context, I think this study demonstrates that, if you really care, it's really important to achieve as much as possible today.

Yes, a major battery breakthrough could change everything, but Harvey's modularization idea could take advantage of such breakthroughs. Thus, both the legacy effect can be countered, while still embracing battery progress.

Even with a breakthrough; however, there are still massive infrastructure obstacles, for instance. Not to mention plain old consumer ignorance -- which is an extremely relevant FACT. BEV fans can talk until they are blue-in-the-face about the obvious superiority of pure electric car, but consumers don't give a crap.

Chasing large batteries today makes a lot of sense in many ways. But most of the science suggests that without a massive change in consumer psychology, current battery technologies will still not be enough to make large battery vehicles the mainstream solution for more than a decade, maybe even two.

Hence, my question is how long do wait before we starting taking the legacy effect of our collection automotive purchases more seriously?

We already know starting today, it will take at least 20+ years to replace the current fleet. Therefore, I believe the topic of the most battery bang for the buck is a very relevant and great question.

how long to replace just half the entire US fleet?
More to the point is how long it takes to replace the vehicles which drive half the annual mileage.  That's about 6 years; annual VMT falls off with vehicle age.  If you could make every model rolling off the line a PHEV-20 starting tomorrow, more than half the annual mileage would be driven in PHEVs by mid-year 2018.

@Roger Pham,
I trust your info and calculations. Looks like the mentioned LiFePO4 battery is even suitable for a HEV using an e-motor up to 50 kW short burst (w/ battery pack size of 1.3 kWh). The larger battery pack will handle the same power burst with less stress (ie heating less).
This battery pack is perhaps the best (ie cheapest) choice at this point in time for the purpose.
Still some issues remain:
- Do the high power/current bursts reduce battery cycle life and by how much?
Recent article here ( stated: "operating a battery at a temperature of 45°C instead of 35°C halves its service life."
The use of ultracap as buffer would significantly reduce battery in and out currents, thus reduce heating and particularly max current spikes that can be very stressful and life shortening.

- If inexpensive ultracaps are available, wouldn't it be more economical to use batteries with twice energy density, and half power density, that cost the same per kilo (kg) - this is very speculative, assumes that power density can be traded for energy density during design.

- In ultracap-battery combination, ultracaps are expected to last 3 or more times longer than batteries. Once battery pack reaches end of life, it can be electronically (also manually) disconnected from the rest of the system and the car works as a non-plug-in HEV, using just ultracaps - no need (and no fear at purchase time) for immediate battery replacement, car is still usable as a HEV.

- Say you have a plug-in hybrid with 8 kWh battery (plus ICE). You park it with SOC near minimum, leave it parked for a month or two, or even just 2 weeks, depends on battery chemistry (say long vacation, or whatever reason). Battery self discharges. If you have a 'permanent' battery, i.e. some 12 or 24 volt, that is supposed to last 3-6 months charged, you might be able to start ICE. But you won't be able to use any electric motor assistance, until you recharge the large (8 kWh) battery pack. Either by connecting it to grid, or running car motor/generator, say of 10 kW, for about 15 min or longer (being stationary, not driving), to reach some minimum SOC of say 25%.

But if you had ultracap-battery combo, you could charge the capacitor from the standard battery (12 volt) in less than 2 min, to 1/3 SOC (ultracap pack say 300 Wh, Min SOC is 1/4 ie 75 Wh, or 1/2 of max voltage for ultracaps, it's a convention to count that way).
So with combo you can drive the car as HEV (using only 300 Wh ultracap module), with full e-motor assistance. You recharge large battery when you get chance, car is always drivable.

So I'd say combo still has advantages over pure battery storage for PHEVs.

Roger Pham

Yes, in theory, combo has advantages. However, complexity might be an issue, since capacitor's voltage drops significantly during discharge, while battery hold voltage much better.

For an HEV, the engine can come on when power demand exceeds about 20 kW in order to preserve the battery. It is in a PHEV that you don't want the engine to come on during all electric mode, due to the emission issue with a cold engine and the waste of fuel in warming up an engine and letting it get cold again during extended electric mode.

A HEV needs the engine to run for most of the time anyway, so no issue in turning on and off the engine frequently when the power demand is within the engine's peak efficiency point in the map, which is around 15-20 kW for an engine sized at 50-60 kW peak power. That's why the gen II Prius only needed a 21-kW battery pack, and the gen III Prius has a 27-kW battery pack. Note the increase in battery power without increase in battery capacity nor battery weight from gen II to gen III. Technology has improved quite rapidly!

An HEV can be designed to start its engine briefly to charge its battery pack whenever the charge drops below a critical point. Just besure that you have enough gasoline inside the tank. But, good point, though, that a larger battery pack of a PHEV will have more loss associated with self-discharge. Thankfully, Lithium Polymer battery has very low rate of self-discharge. I can't be sure about the self-discharge rate of LiFePO4 chemistry, but I'm sure that it's better than lead-acid. Sanyo Eneloop and Sony Cyclelife NiMh batteries also have very low rate of self-discharge.

The comments to this entry are closed.