The utilitarian calculation between the two cases is on almost every accounting identical. To see why, it is important to understand what utilitarianism (or rather its contemporary cousin consequentialism does). In classical utilitarianism, the goal is to maximize happiness. In contemporary usage, "utilitarianism" (used as a synonym for the wordier consequentialism) refers to views that either try to maximize or minimize some quantity and call that morality.
In the usage in the economist, the quantity, if I understand correctly, is the preservation of lives. So, yes, the death of 1 person to save the lives of 5 people is identical in this respect, and thus the utility calculation is identical.
Shifting frames a little -- Mill's Utilitarianism (here the title of his book and his theory) includes a "harm principle" which states that it is wrong to cause harm to others while maximizing utility. It is not entirely clear how Mill justifies this based on other parts of utilitarianism, but it has some interesting consequences for your question.
I think what you are identifying is that it might reflect something about the moral character of the agent in question to be the actual person who pushes someone onto the tracks rather than merely flips a switch. The thing is that most utilitarian accounts, do not think this sort of sentiment is morally valuable. In fact, some utilitarians recommend we go against our intuitions to produce optimal results. Conversely, many non-utilitarians see this inability to incorporate character as a proof that utilitarians are wrong about morality or at least missing something important (Bernard Williams).
Now, I can return to my original qualification. I had said almost every accounting. The problem is that "utilitarianism" can be used as a stand-in for consequentialism, and consequentialism could be about maximizing or minimizing anything as morality. For instance, you could want to maximize justice or minimize harm (Peter Singer) or maximize beautiful music or maximize personal growth or the glory of God. Any of these can be a "utilitarianism", so clearly some of them could see a quantifiable change in character development.
Or even those not directly interested in character development might have a rule against physically pushing people to their deaths even when beneficial (a rule that could be justified or unjustified internally). For instance, you might think that creating people who can casually push another to their death is going to make them depraved such that they will engage in further immorality.
Regarding your last question about social and legal ramifications, these impinge on philosophical views only through philosophical channels... so the differences would only matter if they morally matter here.