"Using the smoking-ban to separate effects, they find a peer effect of 40%! Roughly speaking, this means that the spouse of a person that quits due to a smoking ban is 40% less likely to smoke than otherwise would be the case. That is a huge public health multiplier."
http://www.voxeu.org/index.php?q=node/1182"We find no evidence that higher taxes prevent smoking initiation, but some evidence that higher taxes are associated with increased cessation."
http://papers.nber.org/papers/w14042"The researchers had surveyors approach people on the streets of the Philippines and offer them the opportunity to open a bank account that paid zero interest. The kicker: all the money deposited would be forfeited if, six months after opening the account, the account holder’s urine test showed evidence of smoking...
more than 1 in 10 of the people approached actually signed up for the account!...
It worked surprisingly well, too. The treatment-on-the-treated estimate suggests that about 30 percent of those who opened an account quit smoking because of the account. That’s a higher success rate than is generally seen among those who try to quit smoking using nicotine patches, etc."
http://freakonomics.blogs.nytimes.com/2008/03/10/put-your-money-where-your-butt-is/