Fatalities are more usefully measured per mile not per person. For instance, in many 3rd world countries the roads are notoriously unsafe but the average person covers few enough miles that their road fatality rate per capital looks better.

Note this applies to OP posted above you as well.

No, I'm pretty sure it makes more sense to measure deaths per capita. If poor urban planning made people's commutes 2x longer, would it suddenly be half as bad for someone to die?

If people start driving twice as much and die twice as much, I would not report the roads as being anymore dangerous.

I don't care if a meter of road is dangerous, I care if the amount of driving necessary to live in a city is likely to get me killed. Or if I'm liable to be killed while walking. Deaths/distance is a limited metric that has no bearing on societal cost. Deaths/value is the appropriate cost benefit analysis.

The amount of driving 'necessary' for commuting is zero. You can live in a homeless camp next to your job, or the postage-stamp sized studio most people could afford within walking distance of where the jobs are. You may get hassled by police every once in awhile, but then again seems safer than driving. Perhaps roll yourself in bubble wrap as well and wear a helmet while walking there. I can name quite a few cities where you can wake up in a heroin daze and shit in the streets and be within walking distance of someplace to earn some cash.

Personally 99.9% of driving I do is for pleasure, the lever I have to adjust is miles I drive not the miles the average person drives. On an individual level, each person can directly move their "miles I drive" lever but only barely move the "average miles a person drives" lever. Deaths/capita is a societal measurement that only has limited usefulness to the individual human, who is determining his own behavior. Deaths/mile is more useful as it actually allows me to examine how dangerous the roads are and plan how far I want to commute or pleasure drive accordingly.

Deaths / capita also allows the twisted viewpoint that someone driving a million miles for pleasure gives a very wrong view to the person driving a few miles for necessity. You claim to care about the amount of driving 'necessary' but then use a figure that includes deaths from people driving to the cinema or bar or a road trip to Vegas, which are not necessary. If you only include deaths per capita based on 'necessary' driving it would be a hell of a lot smaller, especially in comparison to the change in deaths per mile.

I'm assuming you're not suggesting the United States is a 3rd world country...

Based on data from the NCSA[1] for the US and SSB[2] for Norway, I get the following for 2019 (before COVID-19 hit):

USA: 36096 fatalities and 3.262e12 miles[3] driven = 1.1 fatalities per 100 million miles driven.

Norway: 108 fatalities and 45.562e9 km driven = 0.38 fatalities per 100 million miles driven.

Might be I did my math wrong. If not, that's still 3x higher than Norway.

[1]: https://www-fars.nhtsa.dot.gov/Trends/TrendsGeneral.aspx

[2]: https://www.ssb.no/transport-og-reiseliv/landtransport/stati... (using 2019 data)

[3]: I'm assuming they use Billion to mean long scale (1e12), otherwise I can't get their table to make sense.

I'm not sure how you interpreted "many third world countries" have notoriously dangerous roads as literally everybody worse than Norway (one of if not the safest in the world) fatalities as being a 3rd world.

I think you also failed to control for race and income, including the vast (per capita) oil production (which Norway taxes the shit out of) that is used towards the benefit of the average Norwegian (Norways is all the way up next to the Emirates in per capita oil production). One study in Chicago, found that those with 'low' economic hardship had nearly ONE THIRD the rate of fatalities [0]. Compound that with those who the study classified 'white' like the relatively homogeneous Norwegians, and you get several percentage points of correction as well. In short, there's a decent argument to be made that a Norwegian dropped into America is almost or just as safe on American roads as they would be in Norway. That is, roads likely aren't 3x safer in Norway, in fact they may be just as safe, so long as you are demographically like a Norwegian.

In short I would say it's probably not the Norwegian road laws or road system that are largely to thank, but rather the fact that Norwegians sit under a gold mine of which is rivaled only by places like the Emirates. And that gold mine is a source of the single greatest improvement in traffic fatality rate in demographic fatality rate, low economic hardship. So yeah, just be born over a gold mine with a silver spoon in your mouth like the Norwegians, why are Americans so lazy?

[0] https://www.ghsa.org/sites/default/files/2021-06/An%20Analys...

I don't need to control for race or income. The point made was that the traffic fatalities in the US is only 0.012% of the population. My point was simply that while it's not a huge percentage in absolute terms, it could be lower.

While I'm not entirely convinced our more restrictive traffic laws aren't part of the equation, I'll agree that socioeconomic factors are probably important.

Thankfully, both countries have a vision for zero traffic fatalities by 2050. Lets hope they make it.