Wednesday, May 18, 2011

US Fleet Fuel Economy

If you take the total vehicle miles traveled estimated by the Federal Highway Administration that I graphed yesterday, divide by the amount of gasoline consumed according to the EIA (after removing the heavy truck mileage that's mainly diesel), you can get an estimate for the average fuel economy of the entire US auto fleet.  This is different than the EPA's estimates for new cars because it represents the actual performance of the entire fleet on the road, new and old.

The graph above shows the numbers through February 2011.  This series continues to disappoint.  If I had my way, people would have paid more attention to the oil price shock in 2005-2008, and the fact that gas is $4/gallon again now even though the economy has barely recovered much at all, and would have started to change behavior and car models.  However, there's no detectable bend in the curve at all.

Here's an old graph from this piece, showing this estimate annually from 1945 to 2005.

You can see that fuel economy drifted up slowly from the forties to the early seventies, barely accelerated at all in response to the first oil shock in 1973, then took off upwards after the second oil shock in 1979, improving several percent per year through the eighties.  Then it went into gradual drift again in the early nineties.

It seems abundantly clear that the US is going to have to improve its fuel economy at a faster rate in the future to cope with rapid increases in emerging market demand in an oil constrained world.  It would have been better if people had done this proactively.  However, it appears that it's going to take a bigger, meaner, oil shock to get people moving in the right direction.  The good thing is that the automakers have used the last few years to put in place a lot more high efficiency hybrids, plugin-hybrids, and electric cars.  Now we just await the shock that's going to get everyone to buy them.  Unfortunately, it's likely to also trigger a recession.  If people won't learn the easy way, then they'll have to learn the hard way...


Gary said...

The recent recession, although coincident with high oil prices, is not as widely perceived as an energy crisis, as was the case in the 70's. Today, it's the housing bust and financial crisis that gets top billing. The economy was also basically in better shape in the 70's, so cars with better fuel economy actually had buyers. The retrenchment today makes new technology less accessible.

jemand said...

I guess it took two oil shocks in the 70s, and maybe this second time gas hits $4 a gallon, we'll see the same effect now. One can hope, at least.

Stephen B. said...

I am actually surprised the 1990s saw much of any increase in fuel economy at all given what I saw people around me buying back then.

Perhaps the continued improvement resulted from all the older cars from the mid 70s and before slowly and continually dying off.

Lars-Eric Bjerke said...


For comparison I checked the corresponding figures for a European country, Sweden. The distances driven are
Cars 64 Gkm/year
Trucks 11 Gkm/year
Busses 1 Gkm/year
This corresponds to 0.13 Gmile/day for a country with a population of 9.4 M. If you had driven as much or little as we do, it would have been 4.3 Gmiles/day (US population 310 M) rather than the 8.2 Gmiles/day. The average mileage for cars in Sweden are:
Petrol 28 miles/gallon
Diesel 35 miles/gallon
The fuel consumption per petrol or diesel car has been constant the last 10 years but the number of diesel cars have increased considerably and is now 50 % of the sales. Among the rest the flexi fuel cars have a fair portion. This means that a US citizen spends 50 % more than a Swede for car transport although the European diesel or petrol prices are 8 $ per gallon.

kjmclark said...

New AP article out today ( points out that people are holding onto their existing cars/trucks about a year longer than before the recession. Combined with cash for clunkers, there's a dearth of used cars out there. Over time, that will push people to buy new and better mileage.

Remember that people think of transportation as one of their mental buckets, including fuel, vehicle cost, and repairs. The price of gas is a constant reminder to find something with better mileage, but OTOH, paying the high price for another car/truck is a big disincentive. They don't give them away. And gas is still cheap compared to the cost of the vehicle. $190 a month at $4 a gallon for 1k mi @ fleet average 21mpg - vs. $250+ a month car payment.

Besides, the marketers are still 'selling' horsepower. They're talking about mileage now, but people still expect more horsepower with a newer car/truck. It's asking a lot of most people to ignore the marketing.

David Guy said...

I guess you forgot that cheap oil is guaranteed in the U.S. constitution. I forget which amendment.

Kenneth D. Worth said...

Thanks, Stuart for another great post. There is, I think, a really easy explanation for the data. The first oil shock of the 70's moved oil from around $2 to around $5 a barrel. It was, like $4 gasoline today, mostly an annoyance. Median takehome pay of $2500 + additional $200 a month in gasoline cost = no large-scale change in behavior. The second shock in '80 was to $40 a barrel. That forced change, much as $8 gasoline would today. I have for a while argued that that's probably what it will take. We could do it today with gasoline taxes but that is political suicide.

Kenneth D. Worth said...

One quibble and one question :)

Quibble: how can the average (purple line) be almost entirely below the bulk of the data points (green lines)?

Question: what can explain the drop in total US oil consumption if fleet economy is barely budging?

Per said...

The lagre part of new cars sold in Sweden now consume 45 miles per gallon in real life driving. In the last five year the average fuel consumtion has gone down greatly for new cars.

Greg said...

Stuart, I think you might be a little pessimistic here. The '73 oil shock was a genuine shock. It took auto makers a year to stop doing their ostrich impressions, and a year to design more efficient vehicles. The earliest that most people could start to buy more-efficient cars was '76.

And then there is the stock-and-flow delay of replacement of the vehicle fleet - Gary's last sentence about recessions reducing investment is relevant here. Finally, there inertia in buying habits to overcome. The upwards kick starting in '78 is about what one would expect.

The sluggish response to the '08 price jump is probably mostly due to the decrease in car replacement -- and, possibly, to perverse effects from "cash for clunkers".

patrickwolff said...

Government regulation (CAFE standards) was the big driver of the move up in the early 1980s in fuel efficiency. The Obama administration pushed through another big increase a few years ago. This will be a big deal when it starts to take effect in a few years. The effect will take several years to roll through as the fleet turns over.

Robert Wilson said...

During the 70's the removal of tetraethyl lead from gasoline in the US adversely affected milage as did addition of smog controls. Circa 1980 my big Ford station wagon got about 10 mpg. It took a while for automobile companies to engineer engines for the new regulations. There was also downsizing and front wheel drive. At one point I had a full sized Buick sedan that was grossly underpowered. They had furnished it with an engine designed for smaller Buicks. It didn't bother me except while entering certain freeway on ramps.

Per said...

More diesel cars is the way to go for US to reduce oil consumtion. Read more on:

yvesT said...

The only thing the US should do, is of course to raise its totally ridiculous (compared to any OECD country) gas tax level, but it won't do it, way too commited towards total economic suicide

Fleet Navigator said...

fuel economy is important to 80 percent of consumers, despite currently low gas prices. The same percentage also said that they support federal fuel economy standards. The numbers have stayed the same or increased slightly over the past decade.