The horizon is not so far as we can see, but as far as we can imagine

Problems with Economics: The Cult of Utility

I first took Economics in Grade 11. We used the standard college textbook at time, written by Lipsey, Steiner, and Sparks.

It started with axioms–the assertions upon which economics are based. The most important axiom  was the following: Humans act to maximize utility.


What is economic utility? It’s generally defined as how useful a product or service is to someone.

How do you find out how much utility an item has?

It is a revealed preference. That is to say, how much utility something has is shown by how much people buy it and how much they are willing to pay for it.

So, Coca-Cola has vast economic utility, for example, but it has less than it used to have in many countries, where people are buying much less of it.

Basically, utility says, “Whatever action people choose to take is the one from which they derive the most usefulness.” This is known as revealed preference.

This is a circular definition; metaphysical in the worst sense. Any action we take is utility maximization. A person can never fail to maximize utility (within their budget), because their actions are what defines the actions’ utility.

(I am fundraising to determine how much I’ll write this year.  If you value my writing, and want more of it, please consider donating.)

People are utility maximizers. Whatever they do is meant to maximize their utility, therefore anything they do is maximizing their utility.

Note that utility is not happiness. It is not health. It is not meaning. It is not pleasure. It is not avoidance of pain.  It is not self-satisfaction. Utility boosters would claim that it includes all these things.

But you can’t measure utility except as price and behaviour. “If they do it, it must have maximal utility for them.”

This is based on another axiom of economics. “Humans are rational.” This axiom has been hit badly in modern economics, but it still underlies most of the models.

Note that utility is always measured as whatever actions people take. This allows no room for self-destructiveness (they must want to be self-destructive!), now does it allow room for not knowing what is good for oneself, for making bad decisions, for doing things which makes one unhappy, and so on.

Utility is not independently measurable. We can (sort of) measure happiness, or pleasure, or meaning, or physical health. We could then ask, “Do these actions actually produce any of these things?”

For example, money produces quite a bit of happiness up to the point where you have everything you need and some security. After that point, it doesn’t produce much compared to other things like having a loving spouse, say.

If we said, “Humans act to increase their happiness,” a common assertion in philosophy, we would have some way of measuring that, albeit roughly.

We cannot measure utility maximization, because utility is always maximized.

This is a statement of belief. “People must do things because they know those things are useful to them in some way, and they must choose that which is most useful.”

Well, no, no they don’t. You don’t need a Ph.D. to know that, you just need to have lived life and seen that people often don’t know what would improve their situations, and, when they do, they often choose not to take those actions, even when they could and often when they wish they would.

Utility maximization doesn’t make sense. Utility itself is a concept with no “useful” definition. Trying to use utility as a concept in the real world leads to circular bullshit: “Well, people are doing these things, therefore, they must be the best things for them even if they don’t seem to be. Who are we to ban trans fats in the face of revealed preferences?”

This leads to all sorts of arguments that take the form of, “Well, yes, we know X has a bad effect on people, but they have chosen X, so, therefore, we should not interfere.” This scales from drinking pop, to borrowing money for college, to activities which have caused global warming.

Actual decision-making, if it is to be rational, depends on an individual knowing their actual objectives re: the decision. Do we want more happiness? Do we want love? Do we want to be healthy? We can measure those things. (On an individual level, yes, even love. I know when I’m in love–though I grant I used to mistake infatuation for love. Still, infatuation was closer to love than not.)

As a society, we probably want to satisfice. We can’t actually measure progress in utility. We can measure progress in happiness, health, childhood mortality, and even in meaning. We can measure infrastructure deficits, and pollution, and global warming, and income mobility.

Utility is best used when you turn it into a basket of “good stuff” and measure those good things.

As for market economies, for capitalist economies, they work best when you take money (price) and you say: “The more people have of this the happier they should be.” As for “utility” maximization as a society, if money is leading to that basket of goods, you try and get everyone past the point where utility starts to sharply decline.

At that point, you work on policies which directly benefit various metrics. Do parks make people happier?  Build park. Does more free time make people happier? Give them more free time. Does less chronic disease make them happier (or less unhappy)? Work on that.

How do you work on those things? You progressively tax people who are above the utility inflection point. If $40K/year is where your inflection point is, you don’t tax people who earn that amount or less, and you increasingly tax people above that.

“Any income above one million has a negligible effect on measured utility (the basket). Therefore, we are taxing you at 95 percent and will spend that money on activities which do increase ‘utility.'”

Because increased inequality decreases various, actually measurable, criteria such as health, happiness, and life-satisfaction, there is even an argument for a simple, 100 percent cutoff point. “Any income beyond this point makes you worse off, and makes society worse off.”

As for utility, as it currently exists, it is a form of Panglossianism. “Whatever a person is doing, that must be what is best for them!” We know that’s bullshit in social terms (this is not the best of all possible worlds, unless it is the only possible world), and we know it’s bullshit in terms of individuals because we all know people (and maybe are people) who could have been far better off if we’d done a few things differently or if society was a little different.

Economics, as a discipline, has some useful things to say, but it was created in an attempt to mimic not just the sciences too closely, but to mimic the certainties of Euclidian geometry and other axiomatic systems too closely.

And there is no such thing as utility in any useful, pragmatic, usable sense.

If you enjoyed this article, and want me to write more, please DONATE or SUBSCRIBE.


Why Do Millenials Love Bernie Sanders?


The American Death Wish


  1. highrpm

    beautifully written policy stuff.

    unfortunately, one must use their reflective mind to process it and turn it into a moral framework for storing in their reflexive mind. and teddy boy cruz & co. have long since finished building their moral frameworks, now operating solely from reflex using george lakoff’s strict father model for the nation-as-family metaphor.

    how does a nation as collective more from an obedience culture to a nurturing culture?

  2. markfromireland

    I suspect that the forthcoming systemic collapse will ensure that economics will go the way of phrenology and astrology.


  3. Ian Welsh

    I prefer to think of it as haruspicy. Every society needs soothsayers, and economists are the equivalent of the guys who used to divine but examining chicken guts.

  4. Shh

    I continue to be amazed when I come across people who understand that assumptions underlie all logical constructs. Everything after the assumption can be “rational” and still utterly useless. So kudos for this piece. I’ve bashed my face beyond recognition on the steely stupidity of unquestioned assumptions to no avail.

    I do take exception to bashing prescientific means of discerning patterns of nature though. The scientific method suffers from unquestioning allegiance at least as much as the dismal science itself. It is important to keep in mind that cognizance itself has evolved not simply since agriculture and writing, but because of them. There is wisdom yet in forgotten methods. As an example, dousing is still relied on to find well water and is far more effective than any other means currently in use.

  5. Erin Gannon

    Good to know, you guys. Good thing I keep my tarot card skills honed. $10/pop at the local.

  6. gnokgnoh

    I think you mean dowsing or divination (divining), unless there is an alternative spelling for it. You threw me for a minute there.

  7. jkn

    Gellner wrote about this in Plough, Sword, and Book, and pointed out that neoliberalism is inherently vacuous because it regards the price of a thing as objectively reflecting its innate value, when price is simply a consequence of the subjective value a culture attaches to something. IOW, “Our culture values everything correctly, because it does.”

  8. gnokgnoh

    As in, “the soldier doused the Afghani with a towel over his face and called it an enhanced interrogation technique.” Whereas, “the farmer used her divining rod to dowse for water, found it, and dug a well.”

    Really, they’re still using divining rods?

  9. gnokgnoh

    The cult of utility (our choices are always that which have the most utility) is a derivative of related social evolutionary theories that insist that humanity and society are on an ever-improving path (we must be improving, if every choice we make has the most utility, both individually and collectively), both of which have built the greater Church of Progress.

    The original sacred texts of the Church can be found on scrolls in the basement of the University of Chicago. It’s liturgy is bound in leather in Jim Cramer’s trunk.

  10. gnokgnoh

    Whoops. “Its liturgy…”

  11. shh


    Oh! the irony 🙂 typos seem endemic to comment sections. Had we been speaking, the conversation may have taken a different turn. And yea, at least out in the American West, most of the ranchers I knew used them (dowsers) religiously, though not in the cause of religion…oddly enough.

  12. Hugh

    I have found most economic terms and concepts wrong, irrelevant, deceptive, or incomplete. Utility certainly ranks high among these. It can not be defined in any meaningful sense. The way it is defined is circular and without any connection to the real world. It can’t be quantified except via spurious measures and so can not be maximized in either the individual or aggregate.

    I see most modern economists as doing the work of Goebbels. It’s not so much about how they are mistaken or why. It’s about who’s paying them and to what end.

  13. Peter*

    Dowsing for water works even for some non-believers and it is a strange phenomenon. I witched for my well in AZ and the driller hit the water table at 200+ft depth. The act of witching is spooky especially when the brass rods start to cross and you can’t stop them. I even tried to counter the ideomotor effect scientists try to use to dismiss dowsing by consciously keeping the rods level and gripping them loosely and tight. Nothing I did seemed to matter, the rods crossed moving 45 degrees and returned to straight ahead when I backed up or continued forward.

  14. Kawaski Steve

    Ian, you wrote: “As for “utility” maximization as a society, if money is leading to that basket of goods, you try and get everyone past the point where utility starts to sharply decline.”

    I can’t understand what this sentence means. I guess you mean you try to arrange for surfeit though that would be a round-about way to say it.

    FWIW, I never studied Economics, but when I started college at 17 I thought I wanted to study Psychology because I thought that was concerned with knowing one’s self. Totally wrong idea. I dropped out and went to India. (Best move ever!)

    But seriously, does Economics even claim that they are concerned with analyzing or maximizing well-being or happiness? Is utility even supposed to correlate with either of those philosophic goals? Do economists even profess to care about anything other than producing and consuming goods? Serious question because I never studied it and wonder if this is parallel to psychology in that they study stuff that is not ultimately very interesting.

  15. Ian Welsh

    Economics claims that market economies tend to maximize utility, yes.

    It means that whatever income stops producing the stuff you’re trying to get (happiness, meaning, health, whatever) is the income you should be shooting at.

    You see this in the stats. 30-40K or whatever will get you almost all the happiness gains of money, after that, the amount of happiness per dollar is a lot less. (Basically when you’ve got food, a decent place to live, some spending money, you’ve got what income can buy in terms of happiness. This varies by country, in the US you might need more for healthcare, say.)

    You’re probably better off, at that point, just using policy to get whatever you want. (So, instead of trying to increase income, just provide universal healthcare, say.)

  16. Perhaps what is needed here is the addition of a bit of probability theory to the theory of Utility. Agreed individuals don’t always get it right, but they probably intend to! So the wisdom of the masses still works.

    As for market economics I would say that all societies are split between Left and Right; the more dependent on the Left and the more independent on the Right. This will remain the case however rich society becomes on average, though possibly more will transfer to the Right as they pass the point you mention where they can satisfy their basic needs. That means the less Government the better and a lower the rate of taxation as a result.

  17. nihil obstet

    I don’t think it’s just that economists think that market economies tend to maximize utility; they support the notion that market economies maximize freedom. Thereby, they tie private riches into moral superiority. Conservatives want hereditary privilege. When the bold-faced overt argument that “some people are just better than others” became too susceptible to revolutionary objections, they switched over to “some people can righteously attain more than others”. The power and privilege of the wealthy thus become the freedom of all. It’s the same old same old, but reified enough for propaganda purposes.

    Thanks for pointing out how wrong this all is.

  18. Billikin

    Isn’t the idea of revealed preference a move towards making the idea of utility more scientific? That is, instead of a nebulous psychological notion we have an operational definition.

    Consider the idea of force in physics. A force is a push or a pull. OK, but Newton put it more precisely, F = ma, force equals mass times acceleration. Acceleration we can measure without reference to force. How about mass? How do we measure it? Normally we measure an object’s mass by applying a force to it. Uh-oh, circularity! Nonetheless we have certain operations that we apply to objects that measure its mass in a replicable way, and we can apply physical theory to such objects, given its measured mass, and get reproducible results.

    One place that I think that economics goes off the rails with utility and preferences is that people’s preferences are non-transitive, and utilities are not. Someone may prefer apples to oranges, oranges to bananas, and bananas to apples. According to utility theory, they should not.

  19. fgd

    Consider the idea of force in physics. A force is a push or a pull. OK, but Newton put it more precisely, F = ma, force equals mass times acceleration. Acceleration we can measure without reference to force. How about mass? How do we measure it? Normally we measure an object’s mass by applying a force to it. Uh-oh, circularity!

    But note that

    F ∝ a

    isn’t really circular, and represented a fundamental change from Aristotle’s

    F ∝ v

  20. stephen

    Closely related is the cult of GDP.

  21. Hugh

    Anybody know how many hundreds of billions are spent on advertising each year throughout the world? I would guess that about 99.99% of it has nothing at all to do with utility. It has to do with catching your attention and selling you a dream. There is almost never any information that tells you much about product A or could be used to make any comparison, let alone a quantitative one, between product A and product B. If you go to the store and actually look at the packages, you may find some product information, but guess what? it wasn’t producers or advertisers who wanted it there. It was government regulation. And even then it often isn’t used, is used wrongly, or isn’t helpful. How does buying a shirt you like the look of but don’t need got anything to do with utility? Fish that is often one thing is sold as something else (more expensive) or a cheap fish is simply renamed to make it more “sexy”. And then there are all the dietary supplements. As the saying goes, mostly what these do for you is give you very expensive piss.
    Hannah Arendt addressed the issue of pseudo-science or scientificality in her work Origins of Totalitarianism, which compared Nazism and Stalinism, hence the references below. But what she has to say about it applies equally well I think to economics:

    “The strong emphasis of totalitarian propaganda on the “scientific” nature of its assertions has been compared to certain advertising techniques which also address themselves to masses. And it is true that the advertising columns of every newspaper show this “scientificality,” by which a manufacturer proves with facts and figures and the help of a “research” department that his is the “best soap in the world.”9 It is also true that there is a certain element of violence in the imaginative exaggerations of publicity men, that behind the assertion that girls who do not use this particular brand of soap may go through life with pimples and without a husband, lies the wild dream of monopoly, the dream that one day the manufacturer of the “only soap that prevents pimples” may have the power to deprive of husbands all girls who do not use his soap. Science in the instances of both business publicity and totalitarian propaganda is obviously only a surrogate for power. The obsession of totalitarian movements with “scientific” proofs ceases once they are in power. The Nazis dismissed even those scholars who were willing to serve them, and the Bolsheviks use the reputation of their scientists for entirely unscientific purposes and force them into the role of charlatans. ”
    Economics is not about the economy or science. It is about power (and wealth) and keeping it in the hands of those who have it. Economics is not a science. And economists don’t need to be forced to come to their spurious conclusions. Wave a few privileges before their eyes and they stumble all over themselves to say what they are paid to say, in suitable pseudo-scientific lingo, of course.

  22. pe

    how does billikin’s revealed preference help ? if they choose bundle a, they must be maximizing.
    Clearly: like if she floats, she must be a witch. Not helping.

  23. Billikin

    fgd: “But note that

    F ∝ a

    isn’t really circular, and represented a fundamental change from Aristotle’s

    F ∝ v”

    The circularity comes from F ∝ m, not F ∝ a. And it is only circularity at the verbal level. Operational definitions provide an escape from the verbal world, by providing sensory motor schemata.

    Yes, Newton and Galileo provided a reformulation of physics, doing away with the Primum Mobile. Newton’s first law is notable for not claiming to explain motion, only changes in motion.

  24. fgd

    Taken in toto and phrased in terms of symmetry principles, Newton’s law can be regarded as having empirical content.

    Basically if you have a self-contained set of objects O1, O2, 03… you can always assign masses and forces to them such that m1v1+m2v2+m3v3… remains constant, and that the forces all sum to zero. There a many possible way that objects could behave that violate these conditions, and only a few that conform to them.

  25. fgd

    Newton’s laws, rather…

  26. I think you make some interesting points points, Ian — many I agree with — but I also think the blanket concept of utility used by many economists is also susceptible to a critique from a somewhat different angle, and I think that bears on your approach as well.

    Unfortunately I don’t have time at the moment to do more than rephrase a comment I used to address a different post, but I still think the bulk of this is relevant:

    I think the difference between producing actual use value and pseudo value through the potential use of deceptive or immoral (but very possibly profitable) business practices is significant. By using a single deceptive shorthand understanding of ‘utility = money’, our economics language obscures some concepts vital to understanding what is really going on in an economy.

    There are two concepts that I think are important: use value and exchange value. (I know Marx used similar terminology, but I haven’t read him, so my thinking about these concepts may be different than his.) However, there is another form of value that I haven’t seen anyone else use that I think is also a critical (and critically different) value: status value. The thing that’s important about status value is that, as a practical matter, it can’t be created, it can only be destroyed. (It also has a lot gender implications that I can’t go into at the moment.)

    As a brief illustration, the price of an ice cream cone is based almost entirely on its use value. The price of a deed to potential oil-producing land in the Yukon is based almost entirely on its exchange value. The price of many luxury consumer goods (like extremely expensive clothing) is based on their status value.

    So my thinking is that every money-based economic system consists of exchanges which are comprised of some mix of use value, exchange value, and status value. It’s possible for an economic system to increase the overall amount of use value being produced, but not the amount of status value, so to properly evaluate the health of an economic system you have to be aware of the relative amount of each value which is actually being produced. Two economies could have identical GNPs, but if one has a much higher amount of activity devoted to the production of status value, it’s a less healthy economy than the other. Excessive transactions based on exchange value can also be a sign of a sick economy and lead to bubbles and subsequent collapses a la the Dutch tulip bubble or the recent tech and real estate bubbles.

    Ian, your alternative to the economic conventional wisdom notes how increased wealth only appears to increase actual happiness up to a certain level, at which point you get sharply diminishing returns. This is important (and I would agree with an egalitarian economic policy that would radically increase income and wealth taxes beyond that point), but I think it’s worthwhile to recognize the way status value becomes a significant driver — probably THE significant driver — of acquisitive behavior beyond your inflection point, because while this behavior is toxic to society in many ways, it isn’t quite as irrational as your post might be seen to imply.

Powered by WordPress & Theme by Anders Norén