It may come as a bit of a surprise: Despite rising gas prices over the past few years, total consumption of highway fuels in the U.S. has actually increased rather than fallen. Some have seized on this phenomenon — prices and consumption rising in tandem — to suggest that changes in gas prices have no discernible effect on how much gas we actually use.

The idea that gas prices have no effect on consumption doesn’t square with economic theory, to put it mildly. And this Excel spreadsheet (courtesy of Charles Komanoff and the ever-informative Todd Litman) sheds some light on what’s really going on. Apparently, even as U.S. gas prices have risen, so have population and GDP. And GDP growth tends to push consumption levels up — in fact, over the short term, gas consumption seems to be far more responsive to changes in GDP than to changes in prices.

The spreadsheet tries to tease apart the two competing forces, and finds that — all else being equal — each 10 percent rise in gas prices is, in fact, accompanied by a 1 percent decline in gasoline sales within a year. Which suggests that had gas prices remained stable over the past few years, consumption would have risen even faster than it did.

It should come as little surprise that over short time horizons substantial gas-price hikes only reduce sales a small amount. People have only so much flexibility to reduce their gas consumption over the short term — a fact that economists have understood for years.

Grist thanks its sponsors. Become one.

But what remains to be seen is whether a sustained increase in gas prices will be accompanied by deeper declines in sales, as people begin to change houses, jobs, or cars to account for higher transportation costs. That’s what economists predict will happen; but we’ll just have to wait to see how well reality matches up with theory.