Two flies cling to the side of a stagecoach as it rolls across the desert, trailing a thick cloud of dust in its wake. One fly looks back. "Wow!" he says. "Look at what we're kickin' up." OK, it's not exactly a knee-slapping punchline, but this is a metaphor, not a joke. A criminologist once told it to me as a way of illustrating how politicians, police chiefs and activists delude themselves about crime policies. They have an idea to fight crime. They put it in practice. Then, if crime drops, they say, "Wow! Look at what I did!" If it doesn't drop, they blame somebody else, complain they didn't have enough funding or just walk away quietly and hope nobody notices. Critics do much the same. If crime rises, it's the fault of that awful politician. If it falls, well, let's talk about something else. These sorts of claims are made all the time. A very popular one is this: Crime fell after former mayor Rudy Giuliani introduced "zero tolerance" policing to New York City. Therefore Rudy Giuliani's zero-tolerance policing caused crime to fall. A couple of years ago, the federal Liberals made this argument: Since we introduced the gun registry, gun crimes have fallen. Therefore the gun registry caused gun crimes to fall. And here's a more recent one about crime in Toronto: former Ontario premier Mike Harris cut welfare and now Toronto is experiencing a rise in gun and gang violence. Therefore Mike Harris's welfare cut caused the rise in crime. It's not hard to spot problems with these claims. New York City's crime drop started long before Rudy Giuliani became mayor. Similarly, the decline in gun crime in Canada had been going on for years before the gun registry was introduced. As for Mike Harris's welfare cuts, they affected all of Ontario, but gang violence is surging only in a few Toronto neighbourhoods. But the bigger problem with claims like these is what the underlying message of that flies-on-a-stagecoach metaphor. It isn't this or that policy that drives broad crime trends. It is broad social trends. Anyone who thinks we can control crime with this or that quick fix is as deluded as the flies on the stagecoach. There's a lot of good evidence to support this view but a new Gallup poll is a striking illustration. The poll -- conducted last fall and released two weeks ago -- asked people in Britain, the United States and Canada if they or anyone in their households had been victims of crime in the past year. The results were astonishingly clear. Two per cent of Britons said someone in their household had been robbed. Two per cent of Americans said the same. So did two per cent of Canadians. When asked if their house had been broken into in the last year, four per cent of Britons said yes. Four per cent of Americans said yes. Four per cent of Canadians said yes. One per cent of British respondents said someone in their household had been sexually assaulted in the last year. One per cent of Americans said the same. So did one per cent of Canadians. The poll did find some differences. Seven per cent of Britons said they had been assaulted, compared to four per cent of Americans and Canadians. Five per cent of Britons said someone in their household had had a car stolen. Three per cent of Canadians and two per cent of Americans said the same. Twenty-two per cent of Britons said they had suffered vandalism, compared to 17 per cent of Canadians and 15 per cent of Americans. Overall, 36 per cent of Britons said they had been victims, compared to 33 per cent of Canadians and 32 per cent of Americans. The survey did not ask about murder -- for the obvious reason that victims cannot answer the phone -- and murder rates in the U.S. are far higher than in Canada and Britain (primarily because they have more guns which makes American violence more lethal). But still the picture drawn by the poll is absolutely clear: Crime rates in all three countries are remarkably similar. And bear in mind that the margin of error on the poll is plus or minus three per cent, which means even many of the minor differences the poll found could be meaningless. And no, this is not a rogue poll. Other international victimization surveys have found only minor differences in crime rates in Canada, the U.S. and Britain. Canada and the U.S. are particularly close together. Stop and think about what that means. Each of these three countries has its own unique criminal justice systems. Each has its own philosophies and policies for dealing with crime. And their approaches often vary enormously: The U.S., for example, incarcerates people at more than six times the rate Canada does. But despite all that, crime rates in all three countries are very similar. The conclusion is obvious: Crime policies don't control crime rates. The broad state of social development does. Nobody wants to hear this, of course, because it means there are no quick fixes and no way to win elections by beating crime. But reality is reality. The flies aren't kicking up the dust, no matter what they think.