Home Home Initiatives Comments Insight Publications Correspondence Search Resources Profiles Upcoming






Economic & Social Policy

Foreign & Military Affairs

Think Tanks

What the Crime Stats Don't Tell You

Dan Gardner

Ottawa Citizen

21 July 2006

Crime statistics released yesterday by Statistics Canada confirm that crime is soaring and Canadian criminal justice policies have failed, in contrast to tough American policies.

But no, wait. It seems the statistics released yesterday actually confirm that crime is falling and Canada's criminal justice policies are working far better than the bloody-minded American approach.

I'm writing this Thursday, so I can only guess how the latest crime stats will be spun, but I'm confident the preceding two conclusions will pop up here and there. I'm also pretty sure who will be spouting which line: The first will come from right-wing types, the second from the left.

One interpretation that likely won't appear anywhere but these pages is that the numbers released yesterday confirm the crime situation is a mixed bag and that criminal justice policies -- any justice policies, anywhere -- actually have very little effect on crime trends. Of course, no one can score political points with this interpretation, which is why the only people who support it are some criminologists no one pays any attention to and me.

StatsCan found that, overall, crime reported to police fell five per cent in 2005. Good news, one might think. But that drop was mainly the result of declines in counterfeiting, break-ins and car theft.

On the most serious crimes, there is bad news: Homicide was up four per cent in 2005, following a 13-per-cent rise the year before. Aggravated assault was up 10 per cent and assaults with weapons were up five per cent.

For those who want to indict the justice system, there's the indictment. Everyone knows the U.S. got tough on crime in the 1990s and crime plummeted to record lows, but here in soft-touch Canada violent crime is soaring. Clearly the system doesn't work and it's time to get tough. Case closed.

Ah, but wait: First, note that most of the rise in homicides was a result of increases in Ontario, which means Toronto and, more specifically, certain neighbourhoods in Toronto. It's a serious problem but it hardly condemns the national justice system. Besides, even with the recent increases, the homicide rate is two per 100,000 people: That's one-third lower than it was 30 years ago and it's low compared to the 5.6 rate in the U.S.

StatsCan's numbers also showed that overall violent crime was flat in 2005. Compare that to the U.S., where murder increased 4.8 per cent and overall violent crime grew by 2.5 per cent in 2005.

And don't forget property crime: It fell six per cent in 2005 and is now the lowest it has been in more than three decades.

So not only does Canada's model cost a small fraction of the American approach, it produces superior results. Case closed, indeed.

Chances are, if the reader is like most human beings, she will look at these contradictory arguments, get annoyed, and come to a conclusion of her own.

One possible conclusion is that you can prove anything with statistics. This is the view of cynics everywhere, including the renowned epistemologist Homer J. Simpson -- which says something about its accuracy.

Or one could ignore the evidence supporting the conclusion contrary to what one is ideologically predisposed to believe and focus intently on the powerful and persuasive evidence supporting one's bias.

Not very satisfactory. Fortunately, there is a third way out: Get more information that may provide a new way of looking at the statistics.

Consider, for example, a StatsCan study done several years ago comparing crime in Canada and the U.S. between 1980 and 2000. The timeframe was important: Those were the years in which the U.S. radically toughened its justice systems and sent its prison population soaring, while Canada, particularly in the 1990s, did something close to the opposite. Quite by accident, the two countries created a continent-wide experiment in justice policies.

The result? Crime trends were found to be remarkably similar in both countries throughout all 20 years.

In a forthcoming book, Franklin Zimring, a renowned criminologist at the University of California, shows that the parallels run deeper still. Homicide rates, for example, have always been much higher in the United States than Canada, but the trend in homicides -- up, down, or flat -- has been remarkably similar in the two countries for more than four decades. A chart of homicide rates from 1960 to 2002 shows two lines rising and falling almost in tandem, like snakes intertwined.

So there is a long history of crime trends in Canada and the U.S. parallelling each other. And those parallels continue to this day.

In both Canada and the U.S., the big crime drops of the 1990s stopped around the turn of the millennium. Over the early years of this century, crime wobbled up or down a little, showing no big trend in either direction.

Then, last month, the FBI reported that while property crime in the U.S. fell in 2005, serious violent crime grew significantly for the first time in many years. Yesterday, StatsCan reported almost precisely the same in Canada.

The parallel is impossible to miss. So is the conclusion: Crime trends are not driven by criminal justice policies.

Just don't expect to hear this from those partisans of the left and right who would prefer to use crime stats to score political points.



Email Article Format for Printing
Home Initiatives Comments Insight Publications Profiles Resources Search Correspondence




Write to us