I was messing around with some spreadsheets, and out popped this little guy:

state elec2

The chart plots the 50 states – minus Hawaii but plus DC – with the price of electricity on the vertical axis and consumption on the horizontal axis. (I left off Hawaii because it’s a serious outlier, with average electricity prices nearly twice as high as any other state.) A simple regression shows that price “explains” about 42 percent of consumption.

My conclusion? Price seems to be a factor in residential electricity consumption. In fact, the same kind of relationship appeared when I ran a couple of regressions using utilities instead of states.

But I don’t want to venture too much further, because things get murky. In statisics, “explains” doesn’t mean “causes” — and I haven’t controlled for income or for local climatic conditions, either of which could have a big influence on response to prices. Plus, 50 data points is a pretty small number for trying to tease out a statistical relationship with a regression; and data that’s aggregated up to the state level may mask important differences that would be apparent in a more granular picture.

Yet even with all those caveats, it hardly seems crazy to think that price affects demand — and that high prices dampen demand. I mean, the relationship is true in some sense for pretty much every consumer good I think of, even including purportedly inelastic goods like gasoline.

All of which may mean that the Northwest’s extremely low electricity prices may be working at odds with conservation. (Idaho has the cheapest residential electricity in the nation; Washington is fourth cheapest; Oregon is 11th; Montana is 16th.) The Northwest has a laudable history of efficiency and conservation programs, and there are still many more market failures to fix. Yet if policymakers are serious about taking the next steps toward a cleaning up the region’s energy portfolio, they should probably take a hard look at raising prices.

All figures in this post refer to 2008 and are calculated from the US Energy Information Administration data found in Table 5A, here.

This post originally appeared at Sightline’s Daily Score blog.