Cool chart showing the changing US murder rate over time

[The Mikado]

I think that there are some people who would be very surprised at the drop of murder in the last 25 years.

[7,052,770]

I think that there are some people who would be very surprised at the drop of murder in the last 25 years.

People who've read Freakonomics won't be.

[Meursault]

Obviously we still need Tuff On Crime police-state policies.

[bedstuy]

This made me curious to look up my local precinct crime stats.  Apparently, my precinct (approx 6 avenues by 20 blocks)  had 71 murders in 1990.  This year we've only had 2, although we're just coming into the prime summer murdering season.  Pretty remarkable.

[ElectionsGuy]

And yet most people think murder is getting worse based on what they see on TV. Its always refreshing to see statistics and graphs like this.

[🐒Gods of Prosperity🔱🐲💸]

what's up with the late 30s?

[patrick1]

what's up with the late 30s?

End of prohibition and unwinding of organized crime + improving economy after depths of GD.

[Beet]

I'm more surprised by the rise from 1900s to the 1930s... previously I thought the century 1860-1960 saw a steady decline in the murder rate.

[Cassius]

I don't buy that whole argument that the legalisation of abortion was what drove the steep decline in crime during the 1990's. After all, if that was the case, surely one would expect to see higher rates of murder in the 1950's, which the chart doesn't seem to show. Indeed, it could be argued, from a certain perspective, that the 1990's drop in murder rates was actually a result of more draconian measures being adopted, at both a state and federal level, to tackle crime, as well as, potentially, efforts to implement laws reducing gun crime, such as the Brady law.

[🐒Gods of Prosperity🔱🐲💸]

what's up with the late 30s?

End of prohibition and unwinding of organized crime + improving economy after depths of GD.

hard to tell any detail from this chart, but I don't see any general relation to economic trends.

[AggregateDemand]

Indeed, it could be argued, from a certain perspective, that the 1990's drop in murder rates was actually a result of more draconian measures being adopted, at both a state and federal level, to tackle crime, as well as, potentially, efforts to implement laws reducing gun crime, such as the Brady law.

No correlation between gun control and violent crime or murder in the US. National Firearms Act 1934 preceded an steep decline, but Gun Control 1968 preceded an era of record violent crime and homicide.

The difference is our resolve to tackle organized crime. Three-strikes, RICO, and mandatory-minimums have created a record number of inmates, some of whom never even committed a violent crime, but it has been effective for keeping organized criminals locked away. FBI publishes crime reports periodically. They associate about 50% of our homicides and violent crimes to gang activity and organized narcotics trade.

Abortion is an interesting theory, but it will eventually have no correlation as abortion rates fall in the US and birth control becomes more prevalent.

[Tetro Kornbluth]

The most interesting recent theory is not that nonsense about abortion, but rather lead

[IceSpear]

I wonder how much the advances in technology and forensics have done to lower the crime rate.

[The Mikado]

I wonder how much the advances in technology and forensics have done to lower the crime rate.

That's a good question.  My initial reaction would be to say that it has a greater effect on the proportion of criminals who end up caught and convicted than on prevention...it certainly has had a very powerful effect on the former.  Criminals in the 18th and 19th century had a far better chance of "getting away with it" than their 20th century brethren.

[Del Tachi]

I wonder how much the advances in technology and forensics have done to lower the crime rate.

This is probably a significant factor, people are less likely to commit crimes that get them locked-up for an entire lifetime if they think they'll get caught.  

[politicus]

I wonder how much the advances in technology and forensics have done to lower the crime rate.

This is probably a significant factor, people are less likely to commit crimes that get them locked-up for an entire lifetime if they think they'll get caught.  

Murder is generally not a calculated crime, most murders are committed "in the heath of the moment".

[politicus]

I think that there are some people who would be very surprised at the drop of murder in the last 25 years.

Any idea why it increases so massively 1902-1906/7?

[🐒Gods of Prosperity🔱🐲💸]

You have to figure that at least some of the early rate changes here are methodological.

[pbrower2a]

what's up with the late 30s?

1. Rapidly-improving social conditions.

2. CCC projects that gave what would have otherwise been wayward kids work, pay, supervision, and meaning.

3. People were less desperate than they had been in the dreariest part (1932-1934) of the Great Depression.

4. Sudden improvement in police work. The FBI turned its focus on persons who violated the federal law proscribing interstate flight to avoid prosecution. Such made state borders less porous for offenders like Clyde Barrow and Bonnie Parker who may have lived in one state and committed their crimes in others.

5. Change in public attitudes toward criminals from semi-heroes to arch-villains. Criminals like John Dillinger and Bruno Hauptmann went from being seen as colorful rogues to arch-villains for killing bread-winners who had done nothing wrong -- and small children.

6. Repeal of Prohibition took away one of the cash cows for violent gangsters who enforced their will and territory through murder.

7. A heavy use of capital punishment for murder. Electric chairs and gas chambers were unusually busy. 

[Mr. Morden]

I wonder how much the advances in technology and forensics have done to lower the crime rate.

This is probably a significant factor, people are less likely to commit crimes that get them locked-up for an entire lifetime if they think they'll get caught.  

http://bloggingheads.tv/videos/14010?in=29:00&out=31:16

[Tender Branson]

Still way to go to reach 1900 levels again (and therefore European levels).

[SPC]

Still way to go to reach 1900 levels again (and therefore European levels).

A bit difficult to achieve when the United States lacks European demographics.

[Filuwaúrdjan]

Still way to go to reach 1900 levels again (and therefore European levels).

A bit difficult to achieve when the United States lacks European demographics.

Please explain exactly what you mean by this

[memphis]

Still way to go to reach 1900 levels again (and therefore European levels).

The 1900 stats are very dubious. Crime reporting wasn't all that consistent in theold days. If a black person was murdered, or lynched, local enforcement didn't always take it seriously. Also, denying that there is a racial element to the high American murder rate is rather counter-productive to anybody who sincerely cares about murders as a real problem. Oddly enough, it's also something only white people do. Black people, the overwhelming majority of whom are not murderers, of course, do not kid themselves about the problems in their communities. They know better than anybody what's going on.
Also, part of the drop in recent years is about improvements in medical treatment. It's only a murder if the victim dies. Trauma units can do amazing things today, and you see a similar reduction in deaths from motor vehicle accidents.

[SPC]

Still way to go to reach 1900 levels again (and therefore European levels).

A bit difficult to achieve when the United States lacks European demographics.

Please explain exactly what you mean by this

In order to minimize the influence of background variables (and consequently make a Type I error in favor of a misguided policy proposal), one should compare people of similar ethnic and socioeconomic background in both jurisdictions, and then determine which jurisdiction has lower homicide rates within each subgroup. One could have varying theories as to why these variations among subgroups exist (whether they be environmental consequences or, gasp, genetic), but to simply pretend they don't exist and persist in making a direct comparison of two disparate jurisdictions is deceitful.