Great Blog Post on Problem with Mainstream Concept of Constant or Increasing Marginal Utility

"Thus, one cannot talk about preferences for risk independently of the real human activity that has risk as one of its attributes."

This article discusses how marginal utility and utility analysis cannot be disaggregated into risk and monetary payoff because utility analysis was devised to discuss preferences of activities (goods and services, etc.), not about preferences of singular attributes of those activities (like risk etc. which can change depending on other attributes or even depending on other activities, and therefore would make little sense to try to reduce to a static mathematical model).

Of course there are other problems even with the Austrian interpretation of static subjective utility (which arguably is the more correct since they invented it), such as the fact that even utility functions for activities can change over time, and, as mentioned in the article, mathematical utility can not be compared across people (which kind of calls into question the breadth of usefulness of utility functions to start with).

I do disagree with some statements in the post. Particularly: "....Answering such questions may be a job for psychologists, but it is not one for economists."

It's that kind of un-wholistic "I'm for me, you are for you" attitude, that we do not need anymore. Much like you can't disaggregate utility into money and risk, you cant disaggregate the study of human behavior into psychology and economics (etc.).