Being a libertarian is good practice for developing this kind of skepticism, as governments often use specious statistics to support their current agenda. Second-hand smoke is one such area, as Crichton mentions. Anti-drug campaigns are also horrible for pulling numbers out of thin air.

It's also good to be skeptical of libertarians denying externalities:-)

As a student of economics, I am quite aware of the phenomena of market failure. However, the question to be asked before imposing a new regulation is not "is there a market failure?. Rather, we should ask "is the new regulation likely to do more good than harm?"

The central insight of the Public Choice school of economics is that governments are composed of people, and like all people they respond to incentives. Public Choice is widely respected in the Economics profession, though it is generally ignored in public debates. Rather, people have a tendency to assume that the government is composed of perfect angels who will optimally address whatever externalities they are tasked with fixing.

There is room for a "good governance" libertarianism, composed solely of economic arguments and with no moral arguments, that would oppose more than 50% of what modern governments do.

Libertarians don't deny externalities; they just recognize that some of them can't be measured.

Just because something can't be accurately measured using current technology doesn't mean it doesn't exist.

Which is what I just said...

We all seem to be in furious agreement then as that was what the parent post to yours also said.