Political assassination attempts are not an effective way to create change, says more than a 100 years’ worth of data

It’s been a fraught election season further complicated by two assassination attempts on presidential candidate Donald Trump. It’s hard to crawl into the mind of an assassin to understand their motivations, but presumably both wanted to effect change.

However, political assassinations are not an effective way to create change, according to Benjamin Jones, a professor at Northwestern University’s Kellogg School of Management.

“Assassinations tend to be chaos agents rather than agents of change,” Jones says.

Jones coauthored a study with economist Benjamin Olken, a professor at MIT, that analyzed 298 political assassination attempts on world leaders since 1875, 59 of which resulted in death. The study is the sequel to another one that looked at what happens to a nation when a leader dies of natural causes. Jones conducted the studies in order to understand the impact a leader has on a nation’s trajectory.

“There’s very little research on the role of leaders as individuals, and it’s difficult to test because leaders try to claim they were responsible for good and [that] the bad was out of their control,” he says.

What the research found

Jones and Olken found that the probability of a successful assassination is about 20%. Moreover, the probability of an assassination attempt increases as the population of a country increases: Doubling the population increases the probability of an assassination attempt each year by 0.35%.

While this may seem like a small figure, it means there’s a 75% greater likelihood of an assassination attempt in a country like the United States versus a country the size of Switzerland, the study found. However, this doesn’t mean that the success rate of assassinations increases.

“There are probably tighter security measures in places where there’s a higher probability of assassination,” Jones says.

He also notes that on the whole, assassinations are ineffective. In autocracies, where a leader has absolute power, a failed assassination tends to see the leader tightening their grip on power. When a successful assassination does occur, it can move an autocracy to a democracy—but again, the success rate is low.

“Often the assassination attempt backfires and leads to the leader becoming more authoritarian,” Jones points out. “It’s chaos because if someone misses they create change, but not in the way they wanted to.”

Assassinations are particularly ineffective in a democracy, Jones says, owing to the system’s checks and balances on power.

“Democracies are robust,” he says. “An assassination doesn’t change much for them. . . . The best way to make a change in a democracy is to vote.”