Well, in the simplest case (when we are not talking being vs. non-being), the utility function is something that you can shift and even multiply by a constant. The only thing that matters for a selfish rational agent which either not considers ceasing to be or ascribes it some utility is ratio of utility differences. You usually maximize expected utility; and you do not care about absolute value, but only about the actions you are going to take. Shifts and multiplication by positive constants do not change any inequality with expectations of utilitiy. And shifts can make negative become positive and vice versa.

Now, if we consider moral questions with variable count of agents, we can find ourselves in a situation where we want to compare being to non-being - and some people implicity ascribe non-being utility zero. Also we can try to find a common scale for the wish intensities different people have. Buddhism with its stopping of reincarnation seems to ascribe negative utility to any form of being before transcending into nirvana. Whether it is better not to be born or to be born into modern world in Africa is a question that can get different answers in Western Europe; now, we can expect that as accurate a description as possible of Western Europe could cause a pharaoh of Egypt say that it is better not to be born than to be born into this scary world.