One of the often-overlooked benefits of declining driving, particularly among the young, has been a rapid reduction in car crash deaths over the past decade. And those safety improvements have probably been helped by falling sales of super-sized pickups and SUVs, along with other promising automotive technology trends.
But during last year’s debate over marijuana legalization in Washington, I heard quite a bit of concern that permissive marijuana laws would reverse the recent declines in crash fatalities. I recall chatting with a well-meaning tow truck driver—a guy who’d seen the aftermath of a lot of terrible crashes—and he was convinced that legalizing pot would just mean more dead kids. Being a parent myself, I found it easy to understand that perspective. Despite a long-term decline in alcohol-related crashes in the state, drunk driving is still a very serious problem—and I can certainly relate to the fear that legalizing another intoxicating substance would boost car crash deaths.
So far, though, the opposite has been true: crash fatalities in the first part of the year seem to have fallen to a new low. See the chart to the right for details—we only have data through April 23, but extrapolating the early year data through the end of the month, we’ve still seen far fewer crashes during the first part of the year than we’ve seen in a while. The long-term trend towards fewer crashes seems to be continuing.
Of course, the numbers are still preliminary, so we shouldn’t read too much into them. And with such a small sample size it’s a bit early to draw firm conclusions. Still, it’s hard to find evidence in the early accident reports that there’s been any uptick in crash fatalities. In fact, based on the long-term vehicle death trends for the state, it looks likely that this has been the safest start to a year since at least 1980; and if the long-term state fatality trends mirror the nation’s, it may be one of the safest starts to the year since the 1950s.
We may have to wait a while to be absolutely sure of the numbers. But if the early trends continue, other states and jurisdictions would be wise to take notice. Marijuana laws are expensive to enforce, and help perpetuate substantial racial and economic disparities. (I’d recommend this documentary for a stark view of the effects of the nation’s drug wars.) So if there’s no compelling public safety reason to continue enforcing marijuana laws, perhaps we’d all be safer, fairer, and richer without them.
Alan
Although as usual the column is well written and makes points I agree with, the headline’s premise is silly, and one not consistent with Sightline’s typical intellectual rigor – it’s a straw man.
There is not much more legal marijuana around than there was before legalization – legal recreational boo is still many months away, arriving in 2014 at the earliest if ever. Therefore, the only people driving under the influence are the same ones that were doing so before – the ones with a supply.
Write this column the same time next year year to be comparing any data of significance. Otherwise, it’s all speculation.
Eric de Place
I’m not sure I follow you, Alan. Recreational marijuana has been legal in Washington State since December 6, 2012. See: http://www.huffingtonpost.com/2012/12/06/legalizing-marijuana-washington-state_n_2249238.html.
Recreational *sales* are still not legal — and won’t be until late this year or early next (right?) but possession is. That said, I don’t know of anyone who’s documented a change in marijuana use since the law went into effect.
VeloBusDriver
Based on what my nose tells me while walking around certain areas of the city, there is plenty of illegally sold marijuana that is available and being consumed. Absent fear of prosecution, consumption of marijuana likely has risen and will rise further once legally sold pot is available. I have no data to back this up, but I’m happy to take wagers against the results from the first few studies. Anybody care to take the other side of that bet? (Or, is that legal?)
Carole
You’re comparing apples and oranges here. There’s no way you can say anything statistically with certainty. At least compare the same time periods — use the first quarter of 2008-2012.
And then go enroll yourself in Stats 101.
Clark Williams-Derry
I’m not sure I understand your point, Carole. First, the chart of 1Q 2008-2013 shows essentially the same thing. Second, I decided to add the data through April just to make it clear that I wasn’t hiding some sort of late uptick in the fatality count.
I feel as if I’ve added the appropriate caveats about the fact that the data are preliminary, and about the small sample size. I could have added other specific caveats — e.g., that you’d have to do a more robust regression analysis, involving more data, and adding data that captures weather, traffic volumes, and more; and also look at the intoxication statistics when they come out. With enough data, it might be possible to tease out some independent effect of marijuana legalization on crash rates.
Still, I think it’s important for people to see that the immediate spike in crash deaths that some people predicted simply hasn’t materialized.
David Bangs
I wonder how many iPhones would be in use if you could not buy one legally.
I’m thinking there will be an uptake in use once there are stores selling it. If not, the stores would go out of business quickly.
Don’t know what will happen with traffic accidents then. It will be interesting to find out.
Steve Erickson
The flip side here is that the likelihood of there being more gun than motor vehicle fatalities in Washington is going up rapidly. In 2010 there were more gun deaths than vehicle accident deaths. I haven’t seen the numbers for 2011 and 2012. Its looking like that will be the new normal. I bring this up only because a common refrain when anyone points out how many people die in this country every year from guns is that far more people die from cars. ‘Taint so anymore.
Clark Williams-Derry
Very good point!
Bill Bradburd
what are bike accident trends in this period?
Clark Williams-Derry
The figures apparently include bike-car collisions, at a minimum. I’m not sure about solo bike crashes — say, a biker who dies in a collision with a stationary object. That’s probably pretty rare. Also, 2012 and 2013 figures are preliminary; that means that some bike fatalities may be reclassified as “non-traffic,” and could be removed from the counts.
The state department of health may have counts somewhere deep in their system, but they don’t seem to publish bike-specific summaries on their website. And besides, there’s usually a year or two delay in reporting. Another data source for very narrow causes of death I’ve used CDC Wonder — http://wonder.cdc.gov/ — but the delays there are even longer.
Amber Hanes-Miller
I think the bottom line here is that the predictions of naysayers didn’t materialize in the first half of the year. I look forward to Clark’s continued coverage of this subject to see what the behaviors and incidences reveal.