I'd like to ask a question.
How does the "Government" know how many miles are driven on the highways? It appears to me that everyone just accepts the numbers put out by the Government without questioning the methodoligy of the numbers acquisition.
I don't "report" my miles driven to the government, do you? If they are guessing using fuel purchased and average fuel milage, then a chance in average fuel milage could throw off their numbers?
I don't say I flat don't believe their numbers, I am just questioning if there is a possibility that their numbers of miles driven could be wrong?

How does the "Government" know how many miles are driven on the highways?
I assume they use the highway gas tax. I assume they have to figure in MPG average.

If you google "Vehicle Miles traveled" there are different methods employed by various agencies, and quite a few exercises comparing and contrasting the different methods.

My purely non-scientific answer is they found the calculation that showed "a billion less miles," whichever one that is, and used it as the basis for their report. The same way the weather man will say rain is good if you need it for crops but bad if you need it for a parade that day.

I wondered the same thing so I checked the EIA gasoline sales data for the period DOT mentioned. Gasoline sales declined by 0.0024%. Therefore one would have to assume that a lot of gas was sold that didn't get used.

Personally, I think its impossible to measure miles driven. The DOT press release also touts a decline in emissions, so I suspect that this was the motivation for what I consider to government info that is right up there with the inflation data. Bogus.

Well, they model it, of course. And decades of using the model makes comparisons possible to view trends even if we all agree that the number is more of an index than a census.

It does not seem to me that "travel" will decline significantly. A high % of trips in automobiles are very short. 40% less than two miles, I hear sometimes? So "Peak automobile travel" maybe, but people are swapping in new modes (bike, foot, transit, telecommute, trip chaining). I highly doubt that a lot people are sitting at home with empty pantries because gas is above $4.

What THAT means is the US needs to better account for other modes of travel. Start better counting transit ridership, start counting bike riders, and carpoolers and then factor those numbers into the travel data.

Actual traffic counts are routine. Two common methods of data collection include:

1. A countbox linked to tubes crossing the road. Usually used on lower volume roads. Can be used to infer traffic composition (truck vs car)
2. Inductance loops buried in the pavement.

I believe the federal VMT data is based something like 4000 (?) count locations.