Yes, this is often a good way of thinking about policy decisions that will have some amount of failure either way. I.e. which type of failure is less bad?
A similar line of thought can be applied to such things as whether/how to consider the non-custodial parent finances for college financial aid for students with divorced parents. I.e. cost of cheating/abuse versus cost of shutting out students whom you would otherwise consider worthy of getting financial aid.