"If policy X has outcome Y, should you support X?" - I've personally been asked this question a number of times, and suspect it comes up commonly enough in political discussion that its worth evaluating in a more general sense.
Firstly, we have to recognize that the question is NOT "if policy X had ONLY outcome Y, would you support X?" That question boils down to "Is outcome Y desirable" (for example "If you could murder a carload of clowns and make our children healthy, except the clowns don't actually die and everything is rainbows, should we kill the clowns?") - to answer this question is to fall prey to a rhetorical trick, though most likely an innocent one on the part of the asker!
The question (fully formed) is "If policy X has desired outcome Y, and given that it will have other outcomes Z, should you support policy X?"
A possibly correct, but rhetorically futile and intellectually lazy answer is "No, non-aggression principle. Lets move on." If the person asking the question accepted NAP absolutely, they wouldn't bother asking. Whether or not you believe the NAP is absolute, you cannot persuade someone of that merely by restating it repeatedly.
A short and more diplomatic answer, but one that is similarly unpersuasive on its own, is "I don't know what Z would actually be, so I would rather side on the presumption of liberty until you can convince ME that the benefit Y is greater than the costs Z"
However, if you have some time to spare and would like to engage the question head-on, there is some deeper digging to be done before the full scope of the hypothetical can be made clear.
(1) Examine the antecedent/protasis: Is X a feasible recommendation?
(2) Examine the consequent/apodosis: Do we actually want Y?
(3) Examine the conditional: Could X actually conceivably cause Y?
(4) Evaluate the other outcomes: What comprises Z (including costs of executing Y)?
(5) Propose alternative solutions: Is X the most efficient way to accomplish Y?
Most well formed thought experiments of this nature will not fail on cases (1)-(3). Usually any discord here will come from a more fundamental disagreement.
It's important to remember that raising the cost or lowering the benefit of an activity will result in less of it, and the contrary will result in more of it. Most laws will achieve to some degree their specifically stated aims - it is the unintended consequences described in (4) that most often dismantle the justification for the policy.
However, even if Z evidently outweighs Y, there is a natural impulse of well-meaning people that "we must do something!" - which is where (5) comes in. Rather than just explaining why a policy will fail, suggest a way in which the voluntary cooperation of free individuals can succeed!