A couple days ago, I wrote that taxing carbon wouldn't necessarily make electric vehicles economically viable. Yesterday, I did an interview with Professor Jeremy Michalek of Carnegie Mellon, who has done research into that very question. The interview has been lightly edited to enhance readability.

MEGAN: So to start with, can you tell me a bit about yourself, and how you got interested in the question of electric vehicles?

JEREMY MICHALEK: Sure. I'm an Associate Professor of Engineering & Public Policy and Mechanical Engineering at Carnegie Mellon University. I've been interested in vehicles since I was a kid growing up near Detroit. I became particularly interested in electric vehicles while working with other researchers at Carnegie Mellon examining their life cycle implications. The mix of issues with transportation, power generation, air emissions, fuel diversification, and policy makes it an exciting topic.

MEGAN Can you walk me through some of those issues? What are the key advantages--and disadvantages--of electric vehicles compared to gasoline-powered cars?

JEREMY MICHALEK: The key problem with "pure" electric vehicles (battery electric vehicles) is that they have limited range and take a long time to refuel.

The convenient aspect is that you can refuel them from home, but then you have a problem when you want to take a longer trip with them. For example, when I visit my family in Michigan, it's a 320-mile trip, about 5 hours. If I were driving a Nissan Leaf, even at the best efficiency (and even if every rest stop had the fastest refueling capabilities) I'd have to stop at nearly every rest stop nearly every hour, to charge for a half hour. This turns my 5 hour trip into more like an 8 hour trip.

The advantage of pure electric vehicles is no gasoline consumption and no tailpipe emissions. But there are still emissions involved in producing electricity and batteries.

The plug-in hybrid electric vehicle is a nice compromise: using electricity for short trips and switching to gasoline when it's needed.

MEGAN: But then there's the added expense of the battery?

JEREMY MICHALEK: True. I shouldn't have said the key problem with BEVs is range and refueling time without also mentioning cost.

If the extra battery is small, the added expense is small, and there's a chance to make up that cost in fuel savings. If the extra battery is large, the added expense is large, but the additional gasoline displaced has diminishing returns. Cost of batteries is the largest barrier to penetration of plug-in vehicles.

MEGAN: I get conflicting stories about whether battery cost is falling a great deal, or only slowly and incrementally improving. Do you have a sense of why it's so hard to tell whether batteries are getting cheaper and better?

JEREMY MICHALEK: There are a lot of reasons.

One is that numbers cited often compare apples and oranges. If someone quotes a kilowatt per hour (kWh) cost for batteries, one should ask whether this is referring to the cells alone, or the entire pack--which includes power electronics, packaging, cooling infrastructure, etc.

The cost also varies depending on the power requirements: smaller battery packs cost more per kWh than larger battery packs, because they need to be designed to deliver more power per cell. Some people even cite things like the costs of laptop batteries, and other batteries that are not adequate for automotive applications.

Finally, the future cost of batteries will depend on a lot of factors, such as the size of the economies of scale we get with higher production volume, how much we learn about how to manufacture batteries more efficiently, and the magnitude of technological advances that could bring costs down. Any projection made about these factors into the future will be uncertain, so I'm skeptical of anyone who offers a single point estimate for future costs.

MEGAN: Do you think it's plausible that we'll get to a battery cost that will make pure electric cars competitive with hybrids or old-fashioned internal combustion engines?

JEREMY MICHALEK: Yes, it's plausible, but that doesn't mean it will happen. The nice thing about small-battery plug-in hybrid electric vehicles (with gasoline backup) is that the battery is relatively small and can potentially pay for itself in fuel cost savings now or in the near future. These vehicles cost a lot less, so we can buy more of them with a given pool of money, and they are more likely to have sustainable market adoption in the near future.

Pure battery electric vehicles are far more dependent on future battery prices dropping to low levels, and we don't yet know if that will happen. Even if it does happen in the future, starting small is likely the best way to get there.

Think of it this way: If you have a 10-mile battery in a plug-in hybrid electric vehicle, then nearly every time you drive the vehicle, you use most of that 10-mile range to displace gasoline. The investment in the battery is well utilized. In contrast, if you have a 40-mile battery, for many shorter trips this investment is nothing but dead weight.

MEGAN: That's really interesting. I wouldn't have thought of it that way. That's what you mean by "diminishing returns" in terms of gasoline displacement?

JEREMY MICHALEK: Yes. The first mile worth of battery capacity added to the vehicle will displace gasoline every time you use it. The 40th mile of capacity will displace gasoline only on trips over 40 miles.

MEGAN: And that's one of the reasons that your research found that even if we priced the negative externalities of gasoline use--like carbon emissions--we still wouldn't end up all driving electric cars?

JEREMY MICHALEK: Yes. Carbon prices over $100 a ton are enough to dramatically change the electricity sector, yet they do little to effect electric vehicle economics. Carbon pricing would affect the price of gasoline, but it would also affect the price of electricity and batteries.

MEGAN: In some places, the price of electricity would presumably go up even faster than the price of gasoline, right?

JEREMY MICHALEK: When you say "even faster," what do you mean? Per mile traveled?

Electricity cost is a small portion of the overall cost of vehicle ownership, but an economy-wide carbon price will affect the cost of all goods that release carbon in their production (and use). This means they increase the price of raw battery materials at mines, the price of shipping those materials and running the factories that produce the batteries.

MEGAN: Is there a carbon price that would make EVs economically viable (without assuming some dramatic fall in the price of batteries?) Or is it like the Red Queen's paradox: you run faster and faster, but never get there?

JEREMY MICHALEK: Yes, but the exact price depends on a lot of factors. For example, what is the future price of gasoline and batteries? How long will batteries last before they need to be replaced? What discount rate do consumers use to weigh higher purchase cost now vs. fuel cost savings in the future? Etc.

Given all of these uncertain factors, it's not appropriate to cite an exact carbon price, but what we do know is that very high prices (we use $140/ton as a high case in our study) do little to change the economics of plug-in vehicles.

Greenhouse gas emissions from personal transportation are significant. But they aren’t the only factor that matters. Reducing oil consumption has its own benefits, and it turns out that air pollution from producing and operating vehicles causes a substantial portion of overall costs.

Looking at all of these together, electric vehicles aren’t necessarily better than today’s gasoline hybrids. Even under optimistic conditions (charging with zero-emissions electricity, assuming the battery lasts the life of the vehicle, etc.) they are only marginally better than hybrids.

MEGAN: And presumably, those optimistic assumptions aren't particularly realistic: we're not all going to be charging with our rooftop solar panels or wind turbines.

JEREMY MICHALEK: Exactly. Even if you have a solar panel on your roof, if you charge your vehicle at night (as most of us would), the electricity generated to charge that vehicle will come from coal in many regions. It's not just about the average electricity in a region – it matters which plants would turn on to meet the extra demand from your plug-in vehicle. In many regions at night, where demand is low, some coal plants turn off. If extra charging demand is added at night, coal plants may be the first to turn back on in response (because they are cheap).

MEGAN: So under realistic assumptions, it gets much worse: you're replacing dirty gasoline with dirty coal power. Even if coal power is more efficient than an internal combustion engine, you just don't get big gains

JEREMY MICHALEK: Realistic assumptions vary from region to region (and time of day), but yes. With the average US grid mix, and including the full life cycle, electric vehicles with large battery packs can cause more overall emissions than today's gasoline hybrids. Under optimistic conditions, they offer some benefits.

When we monetize these benefits, they come out to about $1000 worth of unpriced social benefits (reduction of air emissions and gasoline consumption) over the life of the vehicle under optimistic conditions. Not enough to make up for the cost of the battery back.

On the other hand, low-range plug-in hybrids (with gasoline backup) look good. They offer most of the benefits of pure EVs at lower cost and with fewer infrastructure requirements. Hybrids don't have the barriers to consumer adoption like range limitations and the logistical challenges of parking and charging.

Hybrids do help move us down the learning curves for battery and vehicle production, helping to make pure EVs more economically competitive in the future. Pure EVs offer more gasoline savings per vehicle, but hybrids and plug-in hybrids offer more savings per dollar. If we have limited funds to spend in encouraging adoption, we should focus on the latter, not the former.

MEGAN: That's a pretty important finding, and not one that you hear about very often. At least here in Washington.

JEREMY MICHALEK: We think that current plug-in vehicle subsidies are misaligned with their benefits. The policy assumes that larger batteries are significantly better. Our results suggest that bigger isn't necessarily better -- on average it is worse, and in optimistic conditions larger battery plug-in vehicles are only marginally better than small battery vehicles.

We have discussed these findings on Capitol Hill.MEGAN: If I may ask, what sort of reception did you get? Did you feel as if people were open to hearing that a plug-in hybrid approach was the way to go? Because most of the people I talk to here tend to get excited by the whizzy pure-electric cars like the Tesla

JEREMY MICHALEK: Mostly quite positive. I've found generally that Congressional staffers like to have all of the information, and we've spoken to the non-partisan research groups like the Congressional Budget Office and the Congressional Research Service, who are seeking truth.

What impact that has on policy in the end is still a question. One staffer told me that the problem is if they try to change the existing policy in today's climate, we may lose it all together.

The Tesla is a fine car, by the way. But I don't see a strong case for focusing our taxpayer dollars on this type of car based on its benefits and potential to be a mass-market vehicle. Pure-electric vehicles are perfectly respectable choices for enthusiasts, as commuter vehicles, and for those who simply like them. But when it comes to spending taxpayer dollars, we should be aiming to get the largest reductions in emissions and gasoline consumption that we can.

That means focusing on efficient gasoline vehicles -- including gasoline-powered hybrid vehicles -- as well as low-range plug-in hybrid electric vehicles, like the Toyota Prius Plug-in Hybrid vehicle, which stores enough electricity for 12 miles of electric travel and switches to gasoline for longer trips. Our studies show that vehicles like these actually offer more net air emission and gasoline displacement benefits than electric vehicles with larger batteries, because of the emissions associated with producing the batteries, producing electricity, and carrying around the extra weight of large batteries.

Today’s policies favor larger batteries: The electric vehicle tax credits established in the 2009 stimulus package give larger subsidies for larger battery packs up to $7500 per vehicle, and California’s misleadingly-named “zero-emission vehicle” mandate forces automakers to build vehicles like pure electric vehicles. These policies are consuming limited resources that could achieve more good if they were deployed elsewhere. In fact, the Congressional Budget Office found that electric vehicle subsidies and mandates offer no net benefits, since the federal fuel economy standards allow automakers who sell electric vehicles to sell a corresponding number of low-efficiency vehicles resulting in no change to total fuel consumption.

A future with high gas prices, clean electricity, and cheap, long-lasting, rapidly-charged batteries would benefit pure electric vehicles. But regardless of potential futures, forcing sales of pure electric vehicles now is not the best way to get there. Rather than favoring specific technologies, we can save more gasoline and reduce more emissions at lower cost with a comprehensive focus on the end goals, including incentives that internalize the cost of pollution and oil dependency and policies that target the most cost-effective options.

MEGAN: As I understand it, the four main ways that government can target lower-emissions transportation are tax credits/subsidies, R&D assistance, building supporting infrastructure, and carbon pricing. Can you talk about the right mix of those things? Do we know what the "right" mix is?

JEREMY MICHALEK: First, carbon pricing should be more general, since it's not just about carbon: It's also about gasoline and about air pollution. One of the things that surprised me in our study is the important role of air pollution. I expected it to be a minor portion of the overall social costs of vehicle production and use, but it turns out to still be quite significant. So the fourth approach is more generally to tax externalities, which could include other emissions besides carbon and could include other taxes on gasoline besides carbon taxes.

The foundation of lower-emissions transportation policy should be taxing the damages done by producing and operating vehicles at a rate equal to the damage they cause. This encourages people to drive smaller cars, drive less, or buy more fuel-efficient vehicles, but it leaves the consumers with the freedom to buy what you want (as long as you pay for the damages you cause). Revenue from such a policy could also be used to lower taxes on things we want to encourage (like employment).

R&D is also important, since companies competing in a free market typically under-invest in research (because the benefits of their findings are too often spread among them and their competitors). Government investment in R&D is one of the few ways we could make a technology breakthrough in batteries that would make electric vehicles more affordable.

Infrastructure, in terms of public charging infrastructure, is a very expensive way of saving gasoline. We have a study out this month in Energy Policy estimating that workplace and retail charging infrastructure costs at least $10-20 per gallon of gasoline saved. It's cheaper to buy more batteries or cars.

Finally, tax credits and subsidies can be justified if they help us "get over the hump" into a self-sustaining future of electric vehicles that we would not be able to get over otherwise. The problem is, there is no guarantee we can get there for pure electric vehicles -- many uncertain factors would have to align (cheap long-lasting batteries, expensive gasoline, etc.).

The benefit of targeting a more incremental step with plug-in hybrid electric vehicles is that they cost less now, and can even pay for themselves under some conditions. Studies have shown they are preferred by consumers to pure EVs, they provide more environmental good on average in the near to medium term, and they avoid the need to invest in costly infrastructure that will be mothballed if EVs don't take off. (That infrastructure, by the way, comes with assorted vandalism, obsolescence, and parking logistics issues). And they leave open the option to move to either pure EVs or another technology in the future, if, say, someone comes up with a platinum-free fuel cell that makes fuel cells more competitive.

MEGAN: I should have named a fifth option: financing. The famous (infamous?) loan program that funded Solyndra was supposed to help companies cross that chasm to a viable product. Do you have any thoughts on whether this is necessary? Helpful?

JEREMY MICHALEK: My main comment on financing firms with innovative energy technologies is that we need to be willing to accept some level of failure. If there is no failure, then we're not taking enough risk.