Christopher Caldwell examines why partisans are more supportive of policies when they are carried out by “one of their own”:

Just as likely, the process moves in the opposite direction. Outlooks are driven by allegiances, not vice versa. “When people feel a sense of belonging to a given social group,” the authors write, “they absorb the doctrinal positions that the group advocates.” Perhaps disturbingly, this means partisan shifts need not be driven by political events at all. The right analogy, [Green, Palmquist, and Schickler] suggest, is not a contract between a party and its followers but a religion….Asking about policy preferences is like asking which is the best mother – the one who waits at home with buttered toast or the one who always has an encouraging word? Most people will answer: the best mother is my mother. They may answer the same way if you ask them whether the best president is the one who uses drones against terrorists or the one who does not.

There is something to this. Most issue activists, ideological voters, and high-information voters will presumably react to the same policy in more or less the same way no matter the party or politician identified with it, but they make a point of taking an interest in policy issues for their own reasons that make this sort of partisan bias much less likely. Party activists will react to a policy according to how advantageous and useful it is for their party, and strong partisan voters will tend to follow their lead. For partisan activists and voters, the first question is not, “Is this a good or defensible policy?” The first question is, “Does this help my party on issue X and make it more likely to win elections?” This approach to policy questions doesn’t make much sense to issue activists and voters, for whom advancing the desired/best policy is the most important or perhaps only reason to be involved in political life.

Partisans rely on partisan identification to process and filter information on policy issues, and they use it to gauge how trustworthy different sources of information are. This can often deceive people into supporting policies that don’t serve their interests, or it can at least make them reduce their resistance to them. A proposal from “our” side is considered to be inherently more credible and acceptable than a substantively identical one from “their” side, because partisans give “one of their own” the benefit of the doubt, they often judge the proposal in terms of political advantage rather than on merit, and almost all the incentives in a political coalition point in the direction of deferring to party leaders and rather than opposing them.

To be a partisan is to trust that those leaders are generally working in your best interest. The fact that this is frequently not true doesn’t seem to destroy that trust. The last thing that partisans want to hear is that their trust is misplaced, because this throws their reason for being a partisan into doubt. On the whole, partisans are unmoved by the argument that they are contradicting their principles by supporting a given administration policy, because their partisanship does not come from adherence to a set of principles that can be contradicted.