Play the EPA Number Game!
Earlier this week, the Institute for Energy Research (IER)–for which I’m the resident economist–submitted its formal comment on the EPA/National Highway Traffic Safety Administration’s plans to impose higher fuel economy standards on light-duty trucks.
I think many of you will be surprised at our findings. But rather than just tell you, let me motivate it by having you pick two numbers. (Really, I think it might be more fun for you if you actually go through this little exercise.) But first, some context: The EPA is obviously claiming that the new mandate is justified according to a cost/benefit analysis. They admit that there will be costs from the rules, which they are modeling as the increase in sticker price on new vehicles, since it will be more expensive for manufacturers to design vehicles that get the required fuel efficiency. For example, just in the year 2030 alone, the EPA concedes that the rules will impose costs on car buyers of $35.7 billion (in inflation-adjusted dollars, at that time).
So for the rule to make sense, the EPA is of course claiming that from today’s perspective, the discounted flow of future benefits exceeds the discounted flow of future costs. There are two main sources of these benefits: (1) EPA claims that by reducing fuel consumption and hence emissions from US vehicles, the pace of global climate change will be slowed, thus sparing future people damages. (2) For various reasons, vehicle buyers do not rationally take into account lifetime fuel economy when making their purchases. In other words, in the free market the EPA says that car buyers are effectively leaving money on the table, because they are not willing to pay an extra $1000 for a vehicle that is otherwise identical to a cheaper model, even if the first one will have more than $1000 (in PDV) in reduced fuel costs over its lifetime.
In summary, those are the two main avenues of benefits, through which the new rules will shower net benefits on Americans. So now that you have the context, here are my two questions:
(1) The EPA gives a quantitative estimate of how much lower global mean temperatures will be in the year 2100, as a result of this rule. Take a guess as to what they say. (They express the figure in degrees Centigrade.)
(2) Of the total estimated benefits accruing from this rule, what percentage do you think is due to the consumer irrationality, and what percentage do you think is due to the other benefits (which include the avoided climate damage)? For example, suppose the EPA says that in the year 2030, there will be total benefits of $100 billion (which exceed the projected costs of $35.7 billion). If the EPA thought $1 billion of benefits came from forcing consumers to make profitable investments in vehicle fuel efficiency (which are completely private benefits), and that the remaining $99 billion of benefits came from the externalities of avoided global warming, traffic noise, etc., then your answer to this question would be “1 percent.” So, guess what percentage of the EPA’s stated benefits comes from consumers irrationally ignoring completely private benefits of buying more fuel efficient cars.
Now go read IER’s comment and see what the actual answers are. Were you close?
Holy crap, I wasn’t even close. I though that my guess 0.5ºC was pretty low, not low enough apparently. Then I guess around 10-20% (somewhere in that range), once again, not even close.
The EPA apparently lives in fantasyland where they pull numbers out of their anuses.
JF wrote: The EPA apparently lives in fantasyland where they pull numbers out of their anuses.
I understand your frustration, Joseph, but actually that’s backwards. If they were literally just going to make up the numbers, then they would have said it would prevent 1C warming and spare the US $80 trillion in damages. But since they actually had to tie it to a formal model, they were constrained, and were reduced to the absurdities mentioned.
Yes, I understand if they were just pulling them straight out of their nether regions that they would use far more favorable numbers. However, they also are constrained by having to make this whole thing seem scientific and respectable, so they must use “respected” processes.
Either way, the numbers are pretty much crap that means absolutely nothing to the real world– we are talking about the future of both climate and subjective benefits. No matter how they fancy it up with their “scientific” models, they’re still pulling numbers out of their butt. There is not a single person on this Earth that can know any of these things even if he had a team of the very best working on this everyday for the next 10 years.
Politically on this topic, many people are so ideologically driven that can care less about the costs vs benefits. So, the study is pointless, anyway.
I was really close the delta in temperature and really off on the % due to consumer irrationality. I guessed 0.01 degree C on the former and 10% on the latter.
Heh pretty good David. How did you come up with the 0.01 figure? Did you actually try to backtrack it from the damages you knew they had to be avoiding? Or did you figure, “It has to be at least that big for them to say it with a straight face, but on the other hand how much could regulating light-duty trucks in the US possibly affect the global temperature in the year 2100?”
My initial thought was something on the order of 0.01 degree Celsius but I decided to go with 0.5 degrees out of generosity or insanity. I was completely and so embarrassingly wrong about the percentage number that I shan’t even confess to the specificity of my error in judgement.
The EPA is one of those agencies my friends are terrified that Ron Paul might abolish (if he uses magick to somehow defeat Republican election corruption and actually get elected). The “Environmental” and the “Protection” part of the name apparently mean that it is self-evidently a useful and necessary governmental function. “HOW can ANYONE not want to protect the environment?”
See for example this video: http://www.youtube.com/watch?v=UGL-Ex1CD1c. The “anonymous” member at 8:00 seems to think that the EPA is an axiomatic truth. All humans act. Therefore, the EPA is valid.
I shall be referring people to this blog post.
And representing the layman…I guessed one full degree by 2100 would be what the EPA would have claimed this policy could achieve and I thought 25% of consumers being dumb.
I was shocked to see the real numbers, but trying to guess the numbers first made this much more interesting and fun.
Quoting from your link:
Consumers Massively Dumb?
[D]epending on the parameters, anywhere from 56 to 73 percent of EPA’s claimed “net benefits” from its rule, derive from EPA’s assumption that motorists irrationally fail to understand how much money they could save by buying more fuel-efficient vehicles.
I can’t speak for the EPA’s decision to sell this as a climate-mitigating policy intervention. However, that consumers fail to act as super rational homo economicus types when calculating future savings from fuel efficiency gains is pretty well established in the literature. The most thorough (recent) treatment of the issue that I’ve seen is by Allcott and Wozny (2011): “Gasoline Prices, Fuel Economy, and the Energy Paradox”, available http://bit.ly/zai4lu .
Using a huge dataset (“86 million transactions at auto dealerships and wholesale auctions”), they find very strong evidence that US auto consumers consistently undervalue the (discounted) lifetime savings relative to initial purchase costs — by a ratio of around 7:10.
In fact, they also point out that “misoptimizing” plays a much bigger role than climate effects i.t.o. fuel efficiency benefits…
That said, the A&W paper would also support some of your other objections. (E.g. They argue that “CAFE standards appear to be much more aggressive than can be justified by misoptimization alone”.)
Bob, if you haven’t seen the paper, you should have a look.