To add a bit more perspective, the US is also an outlier amongst developed nations, who are almost all under 2.0. That 3.9 is not 'normal' (though it is something of an improvement on previous rates).
It's a lot more nuanced than that as well with the US. A very small portion of the US population lives in a location with a murder rate at 3.9 or higher, because murder in the US is hyper concentrated (a few areas of Chicago have more murder annually than all of Japan). People like to pretend the US is a wasteland of murder, when in fact the very extreme majority of people in the US live in areas with murder rates comparable to Canada.
Of course urban areas are radically more dangerous when it comes to murder. The murder rates in US cities are drastically higher per capita than in rural and suburban areas, that isn't even remotely up for discussion, it's a fact that stretches back to the beginning of US data on murder.
I'm not sure why you emphasize that murder rates are per capita, that proves the point you're trying to deny. The per capita murder rates in cities are far higher. The very high city per capita murder rates in the US are the reason the US murder rate is so high.
The point being made is that you're cutting out the bad parts in the US in order to make the comparison, but not cutting out the bad parts in the comparitor. Rural murder rate in the US is less than combined rural + urban murder rate elsewhere? Quelle surprise.
Similarly, by cutting out the urban parts of the US in your comparison, you're cutting out 70% of the US population.
Well, given that state breakdown, I think the right conclusion is that there are simply more murders in the US than other developed countries across the board, no matter how you slice it.
Well, I somewhat disagree about that. State-level slicing is a reasonable approach because then the units of comparison are typically more similar in size. And there are several U.S. states (New Hampshíre, Hawaii, Vermont, Maine) that have a lower homicide rate than Belgium, Lichtenstein, Taiwan or Estonia.
(Yes, comparing rather marginal differences.)
However, Venezuela is on a completely different level. Moreover, murder rates in the US are heading downwards; they are half of 1980's numbers. Venezuela is radically up, almost ten times of 1980's numbers.
Imagine you broke down Belgium, etc the same way the US is broken down by state. You'd get even lower murder rates in many areas of Belgium, that would make the safest areas of the US look dangerous by comparison.
If you break down Belgium to parts that are similar in size to U.S. states, you end up with one Belgium.
(However, breaking up Belgium brings to mind some controversies: the motto of the country is "Einigkeit macht stark" or "Unity makes Strength"; after the 2010 election (June 13, 2010) it took 541 days of negotiations to come up with a functional government (sworn in on December 6, 2011).
I had a peruse through http://www.city-data.com earlier, and most of the cities I thought of looking at were higher than that 3.9 - only Portland and NYC (oddly) matched it. Denver was 5ish. Dallas 9ish. SF 5. LA 6. Nashville 6. Boston 8. While I didn't go exhaustive, it was enough to establish the trend. I imagine Salt Lake City would be lower than the national rate, but most of the cities were above it - certainly more population than 1-2%.
Get into the cities perceived as high crime (chicago, philly, dc...) and you get 15s and up - nawlins was 40!
And still.... the figure for Venezuela is 64 (per 100 000 pop.). And that is the whole country. Caracas is much, much worse.
Until late 1980's the Venezuelan murder rate was below 10. It was actually lower than United States. Since then, the U.S. has improved a lot, and Venezuela has collapsed, particularly quickly during Maduro's time but that is merely a consequence of Chavez's "Bolivarian" policies.
(I personally think they are giving a bad name to Simon Bolivar.)
Is murder in other countries not concentrated? I can draw circles over a few areas in Japan and say "these areas here have 100% of murders, the rest of the country has a murder rate of zero".
If you step outside of the major US cities, the murder rate per capita plunges. That variance in the US is far beyond what you see in other developed nations (eg Britain, Sweden, Finland, France, Belgium, Germany, Japan, etc). It's so bad that if you just brought the three dozen worst urban sub-areas down to normal as compared to the rest of the city in question, you'd reduce the total US murder rate more than a full point.
>if you just brought the three dozen worst urban sub-areas down to normal as compared to the rest of the city in question, you'd reduce the total US murder rate more than a full point
>three-dozen
That seems like a lot of areas to say "just". If it's that easy to bring it down, do it in actuality rather than coming up with hypothetical scenarios where the US doesn't have such a high murder rate.
Japan does not have anything even close to what would be considered a high-crime US neighborhood, and the little crime there is fairly even scattered around the country:
