An oldie but goodie from RJ Rummel:
Based on theory and previous results, three hypotheses are posed:
1. Libertarian states have no violence between themselves.
2. The more libertarian two states, the less their mutual violence.
3. The more libertarian a state, the less its foreign violence.
These hypotheses are statistically tested against scaled data on all reported international conflict for 1976 to 1980; and where appropriate, against a list of wars from 1816 to 1974, and of threats and use of force from 1945 to 1965. The three hypotheses are found highly significant. Tests were also made for contiguity as an intervening variable and were negative. Finally, two definitions of “libertarian” are tested, one involving civil liberties plus political rights, the other adding in economic freedom. Both are highly positive, but economic freedom is also found to make a significant added reduction in the level of violence for a state overall or between particular states.
Here’s the link, and this turned into an article in Journal for Conflict Resolution. I think he’s wrong. I think it’s a shame that this argument is cited as an example of libertarian thought in international relations, or at least that it’s still cited as The Libertarian Example. It was good when it came out during the Cold War (in 1983). But it’s soooo Westphalian. Trying to bring Philadelphian sovereignty back into the picture is a tough slog.