Skip to content

In the US, both crime and incarceration are too high

In the Washington Monthly today, Keith Humphreys puts up this chart:

Keith probably doesn't know this, but one of my pet peeves is using homicide as a proxy for violent crime. So I redrew the chart with the same countries but using violent crime rates instead of murder rates:

Guatemala is now an ideal country, while Germany and Canada are lawless!

The United States is still part of the Disastrous group, which is fully deserved. Relative to similar countries our incarceration rate is stratospheric and our violent crime rate is one of the highest:

We overbuilt our prisons in the 1980s and then filled them up by increasing our sentencing guidelines beyond all reason. With violent crime already down by half since then, it's well past time to cut prison sentences in half, tear down half our prisons, and get serious about reducing violent crime by another half.

And keep in mind that while murder gets the headlines, assault and robbery represent the vast majority of violent crime incidents in the US (about 90%). That's what makes streets feel unsafe, and that's where we should be focusing most of our attention.

25 thoughts on “In the US, both crime and incarceration are too high

    1. Eve

      Start making more money weekly. This is valuable part time work for everyone. The best part ,work from the comfort of your house and get paid from $10k-$20k each week . Start today and have your first cash at the end of this week. Visit this article
      for more details.. https://createmaxwealth.blogspot.com

    2. cedichou

      Yes, my thought exactly. I'm curious what is the violent crime that dramatically increases France's rating. And it's not even close, it's like 50% more than the US.

    3. Jasper_in_Boston

      Hard to believe France, Belgium, New Zealand and Australia have more violent crime than the good old USA, and Mexico less?

      And it's really hard to believe Belgium is more violent than South Africa!

      Kevin really ought to re-explain his dislike of using murder as a proxy for violent crime. It seems unsupportable.

  1. middleoftheroaddem

    I am actively involved with a charity located in Guatemala, and visit a couple times per year. The data, and related chart, are misleading at least around Guatemala.

    The idea that Guatemala has a superior criminal justice system to the US, boggles my mind: there are areas of Guatemala City run by gangs almost completely without a police presence. Northern Guatemala is a quasi narco state. Similar, violence, bribery, favoritism and nepotism are common.

    As a reminder, there is a robust flow of refugee's from Guatemala to the US, and not the other way around, who foundational claim is a search for personal security.

  2. beardmaster

    So, the communities with whom I work regularly include a ton of Mexicans and Central Americans. I can say with some degree of expertise that the violent crime rates used in those charts are a result of *significant* underreporting. There are sections of rural Guatemala, for example, which have no policing at all.

  3. paulgottlieb

    Is the implication of these charts that all the violent crimes in Guatemala are homicides? When it comes to atrocious assault, rape, etc.. Guatemala is an earthly paradise. This seems very unlikely--to say the least.

  4. Joseph Harbin

    And keep in mind that while murder gets the headlines, assault and robbery represent the vast majority of violent crime incidents in the US (about 90%). That's what makes streets feel unsafe, and that's where we should be focusing most of our attention.

    We ought to talk about white-collar crime when we talk about crime. A few isolated cases make the news (SBF, Elizabeth Holmes, Bernie Madoff, e.g.) but this whole category of crime never gets mentioned when crime stats are published.

    Most people's odds of getting murdered are, I would bet, a lot lower than being victimized by white-collar criminals. And you can add up all the robberies in all the towns and cities across the country and I'm sure you won't get close to the losses from white-collar crime. Estimates put losses somewhere between $426 billion and $1.7 trillion per year.

    If the average Brink's robbery is $10 million, imagine 135 Brink's robberies every day (or 5 1/2 per hour). That's roughly a half-trillion, the lower end of estimate for white-collar crime.

    If that were on the news every night, I bet we'd do something about it.

    Where to start: Indict Donald Trump, arguably the greatest criminal in the history of the United States.

  5. somebody123

    The violent crime numbers aren’t really comparable across countries, because the definitions aren’t the same. that’s WHY homicide is the standard- it’s pretty obvious and universal when someone has been killed. Belgium is a good example- the homicide rate is 1.7, way below the US’s, but its violent crime is much higher, because their definition of “violent crime” is wider than ours. you can’t compare violent crime rates across jurisdictions unless you’re prepared to do some very fancy statistical adjustments.

  6. Dana Decker

    KD: "one of my pet peeves is using homicide as a proxy for violent crime"

    But violent crime may not be reported (especially in corrupt parts of the world). A homicide is almost certainly going to be reported, or discovered.

    Kevin extolls Guatemala. FWIW, here is Freedom House on that country:

    While Guatemala holds regular elections that are generally free, organized crime and corruption severely impact the functioning of government. Violence and criminal extortion schemes are serious problems, and victims have little recourse to justice. Journalists, activists, and public officials who confront crime, corruption, and other sensitive issues risk attack.

  7. akapneogy

    "We overbuilt our prisons in the 1980s and then filled them up by increasing our sentencing guidelines beyond all reason."

    That's how Say's law works 🙂

  8. Pingback: In the US, both crime and incarceration are too high | Later On

  9. duncancairncross

    Homicide is the same the world over
    Violent Crime is NOT NOT NOT

    In the UK (or her NZ) if I wave my walking stick at you and swear at you that is a "Violent Crime" - Violent assault

    In the USA I would have to cause "actual damage" to your person with my stick for it to be "Violent Assault

    THAT is why you NEED to use Homicide as the comparator across countries and across time

  10. Zephyr

    The violent crime rate graph comparing countries is obviously Cuckoo to anyone who actually knows some of these places, so I would suggest Kevin's chosen metric should only be used within the USA when looking at state or regional crime.

  11. JimFive

    Apart from the violent crime v. murder problem.

    Shouldn't a higher crime rate entail a higher incarceration rate? So the divisions on the graph should be around some sort of diagonal axis, not a square axis.

  12. Pingback: Murder vs. violent crime: Which is more reliable? – Kevin Drum

Comments are closed.