Ronald Bailey digs up data on the question:
[W]ill raining missiles down on Damascus stop Assad from gassing his people in the future? Quantitative research by Michael Horowitz of the Center for Strategic and International Studies and Dan Reiter of Emory University suggests that threatening such aerial attacks works about a third of the time. In their 2001 article “When Does Aerial Bombing Work?” in the Journal of Conflict Resolution, the two define coercion as “a threat to inflict pain on a target if the target does not accede to a demand.” The two identify and analyze all attempts to use air power to coerce countries between 1917 and 1999. “Of our 53 cases of air power coercion, 19 (36 percent) were successes and 34 (64 percent) were failures,” they report.
Horowitz and Reiter define air power coercion a “success” when a target changes its behavior as demanded without being attacked. (“Successful threats are those that do not have to be carried out,” as the economist Thomas Schelling wrote in his 1966 book Arms and Influence.) Clearly, the threat of U.S. air power has failed to dissuade Assad from poison gassing his people. The researchers also count air-power coercion as successful if a target yields shortly after being attacked. With regard to Syria, that kind of “success” is still up in the air, so to speak.