Explaining homicide

British Bobbies - first urban police in the world. How's it going?

I just came across a Wall Street Journal article debunking a poverty and crime theory called "blocked opportunities". Puh-leese! For op-ed writers, debunking crime theories is like stealing lolipops from tots. Simple, unethical and just silly.

Take criminologists (like me) who say routine activity patterns (road networks, travel habits, bar locations) produce higher/lower crime opportunities like crime hotspots. The routine activity theory suggests crime drops by disrupting routine activities and targeting those hotspots with arrests, CPTED, dealing with prolific offenders, etc.

The most comprehensive demonstration of a routine activity approach is Britain's Crime and Disorder Act, 1998.

The Act links crime prevention accountability to municipalities, creates partnerships among relevant agencies, sets prevention goals, and uses multiple strategies to tackle crime hotspots.

Convincingly, since 1998 UK police agencies have been finalists at the International Problem Oriented Policing Award program every year but two. They've won the coveted award eight times.

Some claim routine activity goes beyond street corners to whole countries. They predict Western countries have more goods, more cash for illicit drugs, more things to steal, and will therefore have larger criminal opportunities. This results in higher crime rates than in poorer countries. Voila: poverty doesn't cause crime - routine activities do. The International Crime Victimization Survey, they say, proves it.

It doesn't.

Why? Because that is a logic error called the non sequitur. Consider this:

I constructed this graph from that same survey. It shows the US homicide rate plummeting throughout the 1990s. Canada's rate dips slightly. In the UK, where the Crime and Disorder Act has been in place for the entire period, homicide flatlines and then slightly increases.

You could say this is because crime opportunities between the three countries is worse in the UK. That is unlikely. The US has more gangs, drugs, guns and plenty of crime opportunities.

It is more likely routine activity theory just breaks down at this scale and predicts nothing.

Crime in 2010: Up or down?

The Great Depression - some very bad years

On New Year's night, a car crashed into my neighbors yard across the street. When I arrived to check it out (thankfully, no one was hurt) the driver had already fled from his booze-smelling wreck.

What will that look like in official records? A drunk driver? A car crash? A criminal charge for failing to remain? Damage to property? Truth is it might be one, all, or some combination of any of those.

Official stats depend on how busy the police are, the skill and experience of the investigating officer, whether the drunk driver can be found while he's still drunk, and whether the police are even summoned in the first place (in this case, I dialed 9-1-1). Crime incidents turn into stats (or not) based on a whole bunch of things.

None of that helps answer my most important 2010 crime question - Will crime go up or down? Will the Recession linger and jack up misery? Will the 1990's declines resume as potential offenders age out of their crime-prone years?

To some criminologists, visible social conditions and stories from residents tell the tale. Graffiti, homelessness on the street, and other such perceptions are stories that portend the future.

To others crime "data" alone tell the truth. The evidence-based folks slice and dice numbers and serve them up as proof of this or that. As the car crash shows, stats reveal (or not) different things depending on where they are diced and who is slicing.

So 2010. Better or worse? The Washington Post seems to think it will be better.

Last year Baltimore had a 9% increase in murders. Similar disturbing trends exist in a few other cities. Yet crime rates in many US cities continued the Great American Crime Decline that began in the 1990s.

The FBI Rankings of the 2009 six highest crime cities, in order, are: Camden NJ, St. Louis, Oakland, Detroit, Flint, and New Orleans. The lowest crime ranking cities included Madison, Bellingham, WA and Boulder, CO. Should I choose Boulder over St. Louis?

We teach in SafeGrowth that overall city rankings like this predict nothing about life on the street. We learn how to do a Risk Assessment because it is crime patterns within the neighborhood that matter in everyday life - patterns that can make you safe in a high crime city, (New Haven) or unsafe in a relatively low crime city (Vancouver).

How about the country in which we live? Putting stats ahead of patriotism for a moment, the US and Canadian city homicide rates are interesting:

City crime rates rarely tell the story

City homicide rates 2002
per 100,000

Washington, DC: 45.8
Detroit: 42
Baltimore: 38.3
Memphis: 24.7
Chicago: 22.2
Los Angeles: 17.5
Dallas: 15.8
New York: 7.3
Regina: 4.7
Seattle: 4.5
Saskatoon: 4.4
Sudbury: 4
Portland: 3.9
San Diego: 3.8
Edmonton: 3.5
Vancouver: 3.4
Montreal: 3.4
San Jose: 3.1
Winnipeg: 3
Calgary: 2.6
Toronto: 1.8

Am I unsafe living in US cities? The evidence-based folks might say yes (except if I chose San Jose versus Saskatoon or Portland versus Regina). But Vancouver's Downdown East Side or Toronto's Jane/Finch corridor are no doubt far more dangerous than New York's Greenwich Village or Washington's Georgetown.

Which brings us back to the car crash and whether crime will go up or down. I think it depends on where we live, how engaged local residents are in their neighborhood, and what happens in this economy. In some places it'll no doubt go down. In others, it won't.

I think, when it comes to neighborhood safety, the most valuable New Year's resolution we can adopt is neighborhood responsibility and vigilance: Keeping an eye locally and learning how to prevent crime where we live and work.

Civic engagement will tell the story. Our civic engagement.

Have a happy and safe New Year.