>NHTSA estimates that minimum-drinking-age laws have saved 31,959 lives from 1975 to 2017.

I wonder how many assaults and deaths have happened due to underage drinkers being too worried about going to the police or hospital due to their illegal consumption.

And how many didn't need police-assistance or hospitalization because they were sober?

If there's a perfect solution to ensuring that teens suffer no harm from alcohol, that'd be a neat thing! However broadly repealing prohibitions against teen-drinking wouldn't seem to be such a solution.

And yet, for all intents and purposes, it works fine in Europe to have the drinking age at 16/18. Perhaps even better than in the US! And with fewer drunk driving deaths as a percentage of fatal crashes [0].

Maybe because the driving age is slightly higher, training to become a driver is usually more rigorous, and max BAC levels are around 0.05%.

Another aspect might be cultural — in my experience alcohol is much more taboo and simultaneously more likely to be abused by youth in the US than in Europe. This is probably because many grandmothers will give their grandkids a small glass of wine with dinner or a schnapps for New Years or other stuff like that. This de-stigmatizes the drug, thereby making it less edgy and cool to drink.

I remember being 14 at a wedding in Germany and everyone very jovially encouraging me to have a small schnapps shot, and getting very drunk on absinthe at maybe 16/17 in France. No one really cared, and the lesson of the hangover was plenty to whip me into shape.

[0]: https://www.statista.com/chart/5504/the-worst-countries-in-t...

Might require some investigation and modeling to figure out a good cause. For example, the stats based on drunk-driving incidents compared to total might be misleading as the totals might vary significantly.. we'd probably want something more like [drunk-driving-incidents]-per-[driving-hour] for a starting point. Then probably explore stuff how frequently police catch people and typical consequences, how much folks in an area drink, how people drink (late nights at bars, or earlier at dinner?), if people in an area tend to car-pool, how the roadways might play in (safety features and well lit?; twisty backroads?), etc..

Really seems like there should be research on this sorta thing already, but not immediately seeing one on Google, which is odd..

But, yeah, dunno why folks'd do it in the first place. Seems like simple self-destructive behavior.

---

[This study](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3823314/) seems to use a multi-linear-regression (rather than a causative model), but still looks neat.

Overall, looks like they found that stricter regulations and beer-taxes significantly reduced drunk-driving incidents, while drivers-under-24, higher speed-limits, and unemployment had the opposite effect.

I can endorse what you have written about Germany. I think that the vast majority of people in my country consider it a good strategy for preventing alcohol abuse by their children in teaching them to handle it in a cultivated manner themselves. Not in theory, but in practice. And this needs to be done before they start to drink in secret with their friends.