Utility, Substitution and Demand

2: Utility

In general, if my current choice is A, I might change to alternative B if:

• A becomes more difficult
• B becomes easy (fewer costs associated with it)
• the rewards of A are less appealing
• the rewards of B become more appealing
• an undesirable consequence of A becomes more likely (subjectively probable)
• a desirable consequence of A becomes less likely
• a desirable consequence of B becomes more likely
• ...and so on

Spelt out at length, this becomes a mess. We would like to express all of this in one go, and this is where we bring in the other microeconomic topic: utility.

Utility is just a numerical comparative measure of the subjective desirability of something. By "comparative" I mean that a utility on its own has no meaning: the numbers are only useful as a basis for comparing different options. Saying that there are only three options, whose utilities are 1, 2 and 3, is in effect the same as saying that their utilities are 101, 102 and 103, or for that matter 0.1, 0.2 and 0.3.

The decision principle that goes with this is the Maximisation of Subjective Expected Utility:

• "maximisation" because, for a choice between some set of acts, the one chosen has the highest utility
• "subjective" because the choice depends on the decision-maker's preferences and beliefs, not on reality (hence the economic viability of a whole industry of advertising, which improves subjective perceptions of a product without improving the product)
• "expected" because utilities are calculated using the expectation principle; in other words the utility of an act is a weighted sum of the utilities of all the possible consequences, with the most likely consequences getting the strongest weighting.

Now let's apply this to a specific choice.