Do sanctions work? The most comprehensive study comes from Kim Elliott, Jeffrey Schott, and Gary Hufbauer, fellows at the Institute for International Economics, a think tank with a free-market bent. A sanction "works," in the authors' formula, if it accomplishes the goals identified by U.S. policy-makers at the outset of a sanctions program, like ending apartheid in South Africa or undermining Libya's support of terrorism. Examining 35 U.S. sanctions programs in place since 1973, the study estimates they have succeeded 23 percent of the time.
Our best intelligence estimate is that, for the time being, Iran is not working on building a nuclear weapon. Frequently used coy phrases in the media like "nuclear activities" are references to uranium enrichment to low levels suitable for peaceful applications. Ending that, despite its legitimacy under the NPT as an "inalienable right". is the actual goal identified by U.S. policy-makers at the outset of the sanctions and which is likely to fail.
If sanctions fail better than 3 out of 4 times on average they are even more likely than that to fail in the case of Iran since what they aim to defeat is not a moral offense like apartheid or a dictatorial regime already fighting an active rebellion but a nuclear power program with broad support from the Iranian population. The signs of "success" being cited by the administration are only the damages being caused. There has been no effect on the actual "nuclear activities" other than attempts to shield them from military attack. The result of sanction failure in this case will either be war or Iran resorting to creating a nuclear deterrent which we will have to accept.
The sanctions policy simply isn't worth the risk. We should just aim for the best monitoring agreement we can get and allow Iran to go forward with their nuclear power program including low level enrichment.