Thursday, April 30, 2009

Fuel Efficiency Doesn't Lower Demand, It Raises It

It seems intuitive: Increasing the fuel efficiency of automobiles - or anything else that runs on gas - should lower the demand for oil.

It was with precisely that expectation that Congress enacted the Corporate Average Fuel Economy (CAFE) standards in 1975, following the Arab oil embargo. At the time, US oil imports amounted to a little more than one-third of consumption. Today we import two-thirds. After more than three decades of CAFE standards, heightened environmental awareness, and steady improvements in fuel efficiency and engine technology, America's demand for oil is greater than ever. In 1975, highway fuel consumption amounted to 109 billion gallons, according to the Federal Highway Administration. By 2006 it had climbed to 175 billion.

"It seems obvious that rising efficiency in cars, furnaces, and lawn mowers should, in the aggregate, significantly curb demand for energy," write Peter Huber and Mark Mills in "The Bottomless Well," their perceptive 2005 book on the supply, demand, and pricing of energy. "Sad to say, however . . . efficiency doesn't lower demand, it raises it."

Why? Because improvements in fuel economy effectively make fuel less expensive, and when costs fall, demand tends to rise. As driving has grown cheaper in recent decades, people have done more of it - choosing to drive to work instead of taking the bus, for example, or buying a second car, or moving to a house with a longer commute, or sending the kids to college with cars of their own. Between 1983 and 2001, data from the Energy Information Administration show, the number of annual vehicle-miles driven by the average American household rose from 16,800 vehicle-miles to more than 23,000.

"Efficiency may curtail demand in the short term, for the specific task at hand," Huber and Mills acknowledge. "But its long-term impact is just the opposite. When steam-powered plants, jet turbines, car engines, light bulbs, electric motors, air conditioners, and computers were much less efficient than today, they also consumed much less energy. The more efficient they grew, the more of them we built, and the more we used them - and the more energy they consumed overall."

~Jeff Jacoby, in the Boston Globe,
The Fuel-Efficiency Paradox


At 4/30/2009 7:51 AM, Blogger Unknown said...

You mention the continuous and accelerating suburbanization of the United States as an effect of greater average vehicle-miles per household but I would argue that it is the other way around. High gas mileage vehicles are a boon to all in the middle class but its benefits are not so great as to be a factor explaining the growth of American suburban lifestyles.

Nice post Mark.

At 4/30/2009 9:07 AM, Blogger QT said...


I agree with you. There are lots of other reasons that people choose suburbia...schools, air quality, lower density, lower crime rates.

On supply & demand, would agree that efficiency makes a car cheaper to run and therefore, encourages use although operating costs are only one factor influencing demand. One need also consider capital and maintenance costs. The cost of cars like many other consumer products have come down substantially (MP has had several posts to this effect) and cars have become far more mechanically reliable.

We drive cars because they are highly efficient, convenient, reliable, and affordable.

At 4/30/2009 9:20 AM, Anonymous Anonymous said...

congress is just without any rational thought.

what a tribute to critical thinking.

The issue is that people don't demand gas exactly; they demand miles traveled. the less gas it takes the better.

taken to its extreme, free gasoline or infinitely efficient gas engines would cause people to drive without any economic restraint. Congress apparently concluded that free gas would make people stop driving altogether.

all is lost.

At 4/30/2009 9:21 AM, Blogger bix1951 said...

Thus to reduce consumption of oil we need a carbon tax
the other stuff is a scam

At 4/30/2009 10:46 AM, Blogger Ironman said...

There's a problem with the data - U.S. households have themselves changed from 1983 to 2001, so you're not necessarily looking at a true apples-to-apples comparison.

If it helps, here's a look at the average amount of oil consumed per individual American from January 1982 through March 2008 - you'll see that by and large, oil consumption per day per capita has been nearly constant, ranging between 2.45 and 2.71 gallons per day throughout this period of time. Generally speaking, the core reason Americans consume more oil today is because there are more Americans.

At 4/30/2009 12:05 PM, Blogger happyjuggler0 said...

Nice link ironman, as usual you are a great resource.

Eyeballing your (ironman) graph, it looks to me like gasoline usage in the US follows the "business cycle". Also, one can argue that it follows the price of oil with a lag, thanks to inelastic demand (i.e. it takes a long time to scrap a car).

I will say though that it still looks like it backs up the thesis outlined though. Increased efficiency of gasoline usage ought to result in lower gasoline usage per driver if mileage demand remained constant.

Therefore one can conclude that the deadweight theory of high taxation applies to gasoline too. This really should come as no surprise, that low price increases demand.

At 4/30/2009 1:12 PM, Blogger Ironman said...

happyjuggler0 wrote:

I will say though that it still looks like it backs up the thesis outlined though. Increased efficiency of gasoline usage ought to result in lower gasoline usage per driver if mileage demand remained constant. Perhaps. One would have to dig through the EIA's data on gasoline consumption to better isolate how changes in the fuel efficiency of vehicles has impacted that component of oil consumption.

Another potential outcome is that the greater fuel efficiency of vehicles has indeed driven gasoline consumption lower, but which has been offset by increased production of other products made from petroleum.

You're also absolutely correct regarding oil consumption following the business cycle!

At 4/30/2009 3:01 PM, Blogger Milton Recht said...

There is more going on than the increase in miles driven due to the Rebound Effect.

While people drive more with higher mpg vehicles, they generally do not spend the entire amount of extra disposable income on fuel. The remaining dollars are spent on other items, which need fuel to be produced and transported.

The Kaya Identity is a much more robust way to think about energy consumption in a country and the world because it looks at per capita GDP, energy efficiency to produce one unit of per capita GDP and population growth.

In addition, automobile accident rates correlate with miles driven. The increase in miles increases auto accidents. Higher mpg's result in more accidents. More accidents cause more automobile related injuries and deaths.

The Wikipedia Kaya Identity

At 4/30/2009 5:59 PM, Anonymous Anonymous said...

If total gasoline demand is rising, why are governments calling for mileage taxes because (so they claim) excise tax revenues are decreasing? More fuel used should be increasing excise tax revenue. My sense is that they're lying, they just want a mileage tax as a mechanism to control the population.

At 4/30/2009 7:10 PM, Anonymous Anonymous said...

I think it is called Jevon's Paradox. More efficiency means you get a greater "profit" per use, so you use more.


At 5/01/2009 7:59 AM, Anonymous Anonymous said...

This post completely fails to mention energy efficiency and greater productivity.
Increasing output(miles driven) using less input(gasoline) is the miracle of the free market responding to price signals(higher gas prices) and incentives(fuel efficiency standards).So total demand may be higher but consumption of oil per $ of GDP has fallen 50% since the first oil crisis. Actually according to BP Statistical Review US oil consumption has been the same( around 19million barrels\day) for the last 30 years.

At 5/01/2009 9:10 AM, Anonymous DB said...

"Americans consume more oil today because there are more Americans."
- More to the point; more Americans that can afford to consume energy (Fuel) at the rate that the common efficency allows.

Some questions:
-In America, how does fuel consumption correspond with disposable income?

-Is refining technology a variable in the rate of consumption?

Is there data that reflects regional fuel consumption? It would be interesting to see what regions of the country consume more fuel per capita. The suburban jungles or the dense urban centers.

If fuel efficency is the reason for the increased consumption, the most fuel efficent automobile catagory would show the relationship in sales numbers (or manufactuing data). (?)

Great place by the way.


Post a Comment

<< Home