> But whenever risk compensation has been subjected to empirical scrutiny, the results are usually ambiguous, or the hypothesis fails spectacularly. And when ri

> But whenever risk compensation has been subjected to empirical scrutiny, the results are usually ambiguous, or the hypothesis fails spectacularly. And when risk compensation does play a part in behavior, it tends to do so in small and specific ways... > The question—for driver safety, sexual activity, or public health—isn’t whether some individuals change their behavior in response to perceived risk. It’s whether, at the population level, an intervention makes the world a safer and better place. Just to make it clear how the article is arguing that "risk compensation" is generally a bogus concept.