Proponents of the idea that "life is a precious gift" (i.e., that life is always worth getting) often cite as evidence the fact (based, presumably, on introspection and a quick, imaginary survey) that most people are glad to be alive. Generalized a bit, the underlying argument can be restated as an accurate proxy is a morally acceptable proxy.
On the other hand, situations where a proxy decision is anything but accurate are often defended as moral, for reasons not related to the accuracy of the proxy "prediction." Often, such arguments contain implicit assumptions about the relative value of utility to one being over another.
One such argument is this:
Although I am obviously pro-abortion, this is a pretty stupid argument. It is an argument about privacy - one that would have no problem with New Abortion, which is obviously a moral horror. It is an argument that, specifically, the utility or disutility of a fetus upon coming into existence is so connected to the actions of the proxy (here, the parent) that the utility of the fetus is not a proper concern of anyone but the proxy. That is, the utility or disutility of the possible person does not matter in the greater scheme of things, and must never be used to influence, in the slightest way, the actions of the proxy. It seems that the implicit account of interpersonal utility valuation is this: the caused owes his being to the cause, such that his utility is only of value - positive or negative - to the cause.
I have recently seen this argument applied to circumcision. In the United States, the genital mutilation of female babies is illegal, but the genital mutilation of male babies is widely practiced. The "accurate proxy is a moral proxy" argument obviously fails here, since despite all the arguments for the benefits of circumcision, the rate of adult males choosing voluntary circumcision is lower than even the suicide rate. I have actually seen the knee-jerk liberal bullshit argument for privacy applied here, however - the idea that "if you don't think circumcision is right, don't circumcise your babies." (I think this is more metonymy than actual argument.)
This is also one of the more easily dealt-with oppositions to the antinatalist position - the idea that, if you think it's wrong to have babies, don't have one - leave the rest of us alone! This implies that the well-being of the baby is of no concern to anyone but the potential parent. It is a very strong position to take about interpersonal utility comparison, but it is the only position consistent with the folk argument from privacy.
2. Classical Economics: "You can't create a negative externality by creating new people."
Another place where an implicit theory of interpersonal utility valuation resides is in the definition and treatment of the "externality" in microeconomics. A "negative externality" is a cost imposed on someone without his consent through a voluntary transaction of others, that these others don't pay for. It's the thing that's not allowed in Pareto efficiency: you can't make somebody worse off without compensating him; that's wrong.
However, it is often asserted in the classical economics tradition (to the understandable consternation of mortals) that it's impossible to create a negative externality on existing people just by making new people. That is because any harm the new people do to the old people (e.g., by raising demand, hence prices, for goods) is compensated by the good the new people do to themselves. This contains a very strong implicit statement about the comparison of interpersonal utility: a possible person's utility is just as important as an already-existing person's utility. This is nearly the opposite of the "if you don't like abortion, don't have one" claim, where a possible person's utility is valued at nothing compared to the existing person's utility. Here, they are valued strictly equally.
Obviously, they can't both be true.