In my March 13 blog I had asked if Canada should consider abolishing Daylight Saving Time (DST). The empirical evidence leans more and more against DST. It is bad enough that DST does not generate any significant energy savings. Much worse is that the DST switch in the spring forces the equivalent of a 1-hour jet lag on almost the entire population. Only a few places, such as Saskatchewan and a few municipalites in British Columbia's Peace River region, do not switch to DST.

The ink on my blog wasn't quite dry yet—or whatever metaphor is appropriate for the keyboard and mouse age—when a new research article landed on my desk. My economist colleague Austin Smith of Miami University's Farmer School of Business has just published a brand-new research paper Spring Forward at Your Own Risk: Daylight Saving Time and Fatal Vehicle Crashes, which looks at the 2007 policy change in the United States that extended DST by several weeks. Smith looked at fatal automobile crashes between 2002 and 2011 and found that the transition to DST caused over 30 deaths at a social cost of US$275 million annually. Quite clearly, the one-hour sleep deprivation imposed on us is not good for us.

The paper is quite careful to control for competing explanations of the fatalities, including drowsiness, drunk driving, and bad weather. It is the drowsiness induced by DST that is associated with higher fatalities, whereas DST has no discernible impact on the other causes of fatal accidents. Smith also considers whether it is sleep deprivation or the time-shifting of ambient light that is responsible for the negative effect of DST, and the statistical results clearly point toward sleep deprivation.

Austin Smith's research only looks at fatal vehicle crashes, which is a small percentage of all vehicle accidents. Societal cost of DST switching is thus much higher once one allows for non-fatal injuries and vehicle damage. But Smith's research is very compelling and careful, and I hope that politicians take notice. Why not put DST to a referendum and let voters decide whether to keep DST as is, return to year-round standard time, or move to year-round DST? Perhaps a referendum question that truly affects everyone will help boost voter turnout.

And by the way, another interesting result in Smith's paper is that raising gas prices reduces fatalities. A 10 percent increase in gas prices reduces fatalities by about 0.5 percent. You may think that this is not a large effect, but multiplied with the annual number of road fatalities in the United States—32,675 in 2014—this amounts to about 160 lives saved. If the U.S. needs to fix their fiscal deficit, raising the gasoline tax is among the best available options. Even better, raising the gas tax will save lives.