I’ve found a number of the comments on my post about utility maximization to be helpful and clarifying, and I’m convinced as ever that we need to throw it out. I’m mainly focusing here on posts that either confirm or deny my assertion that it should be thrown out, and leaving out the certainly relevant and helpful comments on why economists rely on it/what it implies.
First, Tom Hickey wrote:
the subliminal worldview of most people is not internally consistent, because it is not consciously constructed or intentional arrived at. A lot of people don’t pursue goals that they would if they had different norms, for example.As a result, most people are conflicted much of the time. They control their “base nature” through the imposition of value-structures, religious and moral codes, ethical standards, and positive, for example, and this is an internal conflict that manifests in decision-making. Some of these decisions are economic choices that are not necessarily utility maximizing in the rather simplistic way that many economists consider, e.g., pursuit of self-interest.
Kevin disagreed with both Tom and me:
On the other hand, assuming someone doesn’t maximize over preferences seems very, very strange. Surely we are guided by passions, but we are not machines. To the extent that we have what philosophers call “human agency”, then we are attempting to do what we think is best for us. That simple idea turns out to be a very powerful paradigm of the world…
Social science is too complicated to know *exactly* why something happened. As such, our models will, by definition, be wrong sometimes. A better criteria for models is whether they are useful!
Note further that max u does not require people to do what’s “in their best interest”: as an economist, I don’t even know what term could mean. I only care that they have some preferences and maximize over them, harmfully or beneficially.
Even if you find that restriction too strong, max u may still be salvaged to the extent that most economic models only care that people act “as if” they maximize utility…I think it’s madness to say that humans aren’t purposeful. It is very, very easy to get people to react to incentives
David Youndberg jumped in:
I’m surprised you say “it’s hard to argue that individuals are maximizing anything.” People are maximizing things all the time! They buy more during sales…
Does this risk being tautological? Perhaps, but if you want to expand the number of things economics explains, that’s a natural danger.
This fallacy, that humans are so purposeful, is what is woefully misguided. This may be a cheap shot, but I think the main reason it persists is because people who are attracted to economics tend to act so “purposefully” themselves…
I meant to say, they aren’t consistently maximizing anything, at least not in a measurable way…The tradeoff of simplicity for a tractable model is not a worthwhile one when the model turns out to do an awful job of predicting how the economy actually works. Yet economists cling to it…
I was too strong when I said humans aren’t purposeful. I meant they are not consistently purposeful. They also consider a host of changing norms and ideals that cannot be modeled. These things can neither be measured (at least right now) nor assumed away.
Finally, I think Sandwichman really brings it to the table. His comment is the sort of thing I would like to write if I understood these things much better:
The gestalt theory of figure and ground is apt here. What counts as figure and what counts as ground is a cognitive choice the mind makes based on subjective experience. Maximizing utility is something that only occurs after we’ve identified what we think the figure is. The illustration of Rubin’s vase demonstrates what happens when the information is ambiguous and could be read either way. It’s also a simplification in that the ambiguity is only two-way. There can also be multi-layered ambiguity. This is what people mean when they talk about uncertainty.
Kevin presents a very articulate defense of utility maximization that sidesteps the fundamental nature of uncertainty. It’s not a matter of satisficing meeting the criteria of maximization, it’s a matter of not knowing and not being capable of knowing what one’s preferences are. The problem with neoclassical economics is that economists have picked some trivial examples of consumer preference and generalized from them. It seems intuitive if you’re talking about preferences between, say, beans and brocolli. The whole preference ordering game works there only because… it doesn’t really matter. For all intents and purposes it’s “reversable”.
On big ticket items, though (things like career, life partner, domicile), the distinction between figure and ground is fundamentally uncertain. Economists have swept that huge difficulty under the rug by treating these as a “normal goods” the preferences for which can be modeled by generalizing from trivial and totally dissimilar examples of commodity preferences. Even the choice between war and peace can be trivialized to choosing between guns and butter.