I was recently talking with a guy in the Econ. Ph.D. program at GMU (which I am now also a member of :D).
He was discussing utility maximization subject to constraint. After some discussion he began talking about the possibility of negative utility. I stated that the minimum level of utility is 0 because if some consumer is faced with two utility-negative consumption choices they will select to consume neither.
The guy said that I should consider that if the consumer has to choose one of these baskets he will select which ever is less negative. While I am in some sense inclined to agree, I am also in some sense inclined not to agree. First, in this situation there is necessarily no free market. It also got me thinking about the idea of utility in general.
Basically, I want to claim that if a consumer has one choice and only one choice then the utility derived from that choice is in some sense 0. It may not be 0 in the sense that they may be more or less happy than a previous point in time, but it is 0 in the economic sense including opportunity cost. I would call this the baseline utility level.
This is utility in the non-remediable sense, or including opportunity cost. It’s like, “zero utility compared to what?”
Which brings up the next point. Utility is always positive using this approach because if Ua<Ub then a is the baseline and vice versa. Negative values become impossible under this approach. Although you could have many goods equal at the baseline.
I think this approach is more rigorous then the previous approach which was to let an individual arbitrarily describe the utility they gain from something. This has known cognitive bias problems. Once we create this objective baseline level of utility, which is the opportunistic utility level, we can then use real values to compare preference for all other goods.
This could provide additional support for a purely objective system of utility measurement through the combination of baseline utility and relative preference in real terms. Particularly revealed rather than stated preference.
Summary theorem: For any consumptive choice to have nonzero utility there must be another consumptive choice with zero utility.
Corollary: If only one consumptive choice exists, it has 0 utility.
Corollary: Let there be some consumptive choice which minimizes utility. Given that this utility level is described as 0, there will be no consumptive choice which results in a negative level of utility.