[aszepieniec]

Gun control advocates cite statistics of the kind "how many deaths are caused by guns" as justification for more gun control. Frankly, whatever the statistic is, it is too high. But that is not the point.

But in order to successfully make an argument along the lines of "more gun control will save lives", they must also cite statistics of the kind "how many lives are saved by guns". Obviously, if guns cause the death of 12000 people yearly and save the lives of 100000 people yearly, gun control is a bad idea. On the other hand, if they only save the lives of 100 people yearly, it might be a good idea after all. I have yet to see a single article on the internet citing such a number. Why is it so hard to find such a statistic?

The number of lives saved by guns (not counting police and military guns) is quite impossible to count. One might succeed in counting the number of cases where guns are used in self-defense, even though these cases usually go unnoticed. But how do you count the number of cases where the would-be perpetrator chooses not to commit the crime on account of a gun being present in defense of the would-be victim? Or doesn't even start to think about the crime?

This is a classic case of Bastiat's "that which is not seen". Because the statistic cannot be calculated, people do not take it into account. But they should if they want to make an effective argument for more gun control based on statistics.

Because the relevant statistics cannot be calculated, statistics should not be used in policy debates such as the one about gun control. Rather, let us reason about the unseen consequences of the actions of individuals as well as of the government. Without involving emotional arguments from either side.

[manicdee]

This argument in support of "lives saved by guns" is not a classic case of Bastiat's "that which is not seen". This is a classic case of intellectual onanism.

There is a high statistical correlation between gun ownership as a percentage of population and gun deaths as a percentage of population. Unknowable figures such as "lives saved by firearms" are already taken into account. Regardless of how many lives have been saved by potential criminals not committing a crime due to fear of consequences, there were far more deaths actually caused by firearms.

This is why the number of deaths got larger: there were more people dead from firearms than people still alive from firearms.

[vacri]

It's also a classic case of selective reporting - you want to include all these invisible cases of defensive gun uses that don't get reported, but you're not mentioning criminal uses of firearms - muggings, holdups, gang activity. If you hold up a store with a gun, it doesn't necessarily follow that you must kill someone, so that stat doesn't get much press - in exactly the same way that defensive gun use doesn't get much press. Well... accurate press. It gets plenty of emotive press.

As for looking for such a statistic, you can start with Kleck in the early 90s, but you should also read criticisms of his - it was a shoddy piece of work with a wide number of fundamental flaws, but it was heavily used by gun advocates at the time. It put 'defensive gun uses' at 2.5M/year and 'lives saved' at 200k/year. The latter number has absolutely no realistic merit when compared against contemporary countries (it would mean that without guns the US murder rate would rise from 4.2 to ~50. For comparison, the current highest homicide rate of any first-world, industrialised nation is 4.2 (which itself is currently triple its contemporaries), and only four countries in the entire world have a rate higher than 50.

Hopefully the above paragraph would help suggest why saying 100k lives saved/year is also extremely unlikely. At 100k you're looking at a murder rate of about 25. Mexico with all its poverty, social troubles, and drug wars only manages 17. Does anyone honestly think that the social fabric of the US is so abysmally weak as to collapse that badly in the event of gun reform?