Taehee Whang, "Playing to the Home Crowd? Symbolic Use of Economic Sanctions in the United States." International Studies Quarterly 55, 3 (2011): 787-801.
Why do we observe economic sanctions despite strong doubts regarding their effectiveness? While the symbolic use of sanctions is advanced as an alternative to the instrumental use explanation, no one has assessed this alternative explanation empirically. I investigate the symbolic use of sanctions for domestic political gain in the United States, assessing in particular the effect of sanctions imposition on US presidential approval ratings. Findings suggest that policymakers benefit from imposing sanctions through increased domestic support. This domestic political gain can present policymakers with an incentive to use sanctions as a low-cost way of displaying strong leadership during international conflicts.
I don't find this convincing. Take the example of Cuba, which as far as I know is currently the longest continuous application of U.S. sanctions. They are related to domestic opinion, but only in Florida, so national presidential approval is irrelevant. This argument might have held in the 1960s, but not really now.
In addition, I would argue that at least after the Cold War, sanctions against Latin America have no impact on presidential approval at all. Take Honduras in 2009, for example. It was a huge deal in Latin America, but barely registered in the United States. All of the debate about sanctions the Obama administration was applying did not impact his approval because Americans did not care about the outcome. Actually, the main domestic audience was a group of Florida lawmakers, and they opposed any sanctions so the presidential approval issue actually worked in reverse.