Foreword: This is a response to Nick Kraftt at Open Economics, where he follows up on his blog in which he “attacks” U Max or “old fashioned economics” by Parsing the Comments that resulted. For non-economists reading this, what follows is my attempt to draw a distinction between utility maximization, which is what economists have tended to assume best describes consumer behavior, and preference maximization, which is more likely to be what consumers are actually doing when markets fail, information is faulty, prices are distorted away from marginal social cost, or uncertainty (i.e., risk) cannot be or is not managed efficiently. We have tended to use the two terms as synonymous. I think the distinction is important. If preferences do not map accurately to something that can objectively be called utility, there is no guarantee that market allocations are efficient in any meaningful sense of that term. (Obey, if you’re out there, this one’s for you. )
Uncertainty, norms, institutions all matter and will shape consumer preferences (more about this later). The real issue IMHO is whether or not individual preferences map accurately to something that could plausibly and objectively be regarded as individual utility. It’s JS Mill’s “better to be a human dissatisfied than a pig satisfied” problem. Mill implies that there is some objective state of the world that corresponds or “should” correspond to human happiness. Were we rational, it is the state of the world we would prefer to that characterized by satisfaction of irrational, short-term pleasures.
We economists have tended to ignore the normative aspect and assume that if someone is a “pig satisfied,” his/her preferences correspond to the highest utility that person can or would want to attain subject to the constraints of their income and prevailing prices (even if there is another feasible to attain (with a little rearranging of resources) state of the world in which the person could have health insurance, retirement security, and better nutrition). We also assume that in forming those preferences, the individual is 1) rational (defined rather narrowly as having preferences that don’t change or switch around at least in the short run and that are characterized by more always and everywhere being preferred to less); 2) self-interested (maximizing his/her own idea of what is best for him/her); and 3) fully-informed (s/he knows everything (and I mean everything) about the good and bad effects of consuming some good, the side effects to others of consuming or producing the good, and the future states of the world that will result from consuming or producing the good. And, finally, the prices s/he faces are assumed to reflect the marginal social cost of the goods and services s/he purchases. If any of these assumptions do not hold, then the consumer’s preferences most likely do not map directly and accurately to a state that could be characterized by JS Mill as a “human satisfied”, i.e., a state in which individual utility by some objective standard is being maximized.
When the assumptions do not hold, the consumer is a preference maximizer, not a utility maximizer. The effect will be to distort demand for things that maximize (uninformed, possibly irrational, maybe even bad in the long run) preferences (e.g., granite counter tops and large homes with en suite baths purchased with no money down and pick-your-payment mortgages) at the expense (opportunity cost) of things that would actually improve the long-term prospects and lives of both individual consumers, the economy, and society generally. In the extreme, we have a world of “pigs satisfied” and an economy with significant (inefficient) distortions in investments in and stocks of financial, human, and health capital.
Norms and institutions are important aspects of market exchange and individual behavior that should (there’s that normative word again) act to align individual preferences with plausibly objective individual and societal utility. In so-called “free markets,” they will embody and reinforce the “social” virtues of generosity, beneficence, fairness or justice as well as individual virtues such as prudence and temperance that lead individuals to take broader and longer-term views of their own behavior and its impact on their own long-term objectives as well as on their community and the larger society.The result is that norms and institutions provide a moral counterweight to greed and other strong incentives to pursue short-term hedonistic “self-interested” objectives.
All policy solutions aimed at remedying or minimizing the divergence of preferences from utility will necessarily involve “paternalism” to a greater or lesser extent. The solution that is least paternalistic and that allows the most consumer autonomy is preferred. To minimize paternalism, it must include improved (and state-funded) education, improved consumer information, increased market transparency, some regulation, and some “nudging” in circumstances where uncertainty makes rationally, informed choice difficult. The more effective are societal norms and institutions at reinforcing virtues that promote long-term individual and societal well-being, i.e., at aligning individual preferences with individual and societal utility, the less paternalism will be required. For this reason alone, prevailing norms and beliefs in support of unfettered self-interest that are derived from misunderstanding the metaphor of Adam Smith’s “invisible hand” and “self-interest” must be corrected.
Assuming virtue in economic models of individual decision making is much the same as assuming full-information. If the assumption is true, revealed preference probably approximates utility maximization. If it is false, then I believe we’re in a second best world.
If individual virtue tempers our “piggy” desires and conditions our choices to something that is both individually and socially better, then the economic rewards of virtue as embodied in and promoted by societal norms and institutions are far greater than we have ever suspected. As economists, we would do well to recognize this when we teach U max.
At an operational/policy level, I think Maxine pretty much nails it. At the theoretical level, I’d hedge a bit on the preference maximization piece, referring you to Sandwichman’s comments that I reposted here.