Utility means usefulness. It explains why you use one thing rather than another. Dis-utility however, means some annoyance, aggravation, discomfort, or regret that you feel or may feel. There is no necessary connection between the two.
Consider the Justice system. David Hume thought it should be useful. After all, it costs money. This gives rise to a Utilitarian approach to Justice- the other side of the coin of which is total factor Productivity. If this rises as a result of a law or a legal judgment, utility has risen. Other jurisdictions will do well to implement something similar. As things are, an avenue has been created through which more and more transactions will be done. Lawyers involved will get more money. But so will their clients. The thing is 'win-win'.
If Utility is considered to be like hunger or thirst, then as more of a thing is available, there are likely to be diminishing returns- i.e. the increase in utility tends to get smaller and smaller. This is why after I've had my third burger, I feel full. If I get a fourth, I may give it away to anyone who wants it. Does money have 'diminishing marginal utility'? If so, the rich won't mind handing over some of their cash to the poor. Indeed, they may favour 'progressive taxation'. The problem is 'dis-utility'. They may be annoyed to have to do so and they may worry that they will regret doing so. Thus, to minimize dis-utility they may put themselves out of the reach of the tax-man. After all, if they really want to, they can always give away the larger accumulated sum at a later time.
It would be nice if utility maximization and disutility minimization were 'duals' and thus some canonical representation of them were possible. Sadly, they correspond to different notions- viz. satisfaction vs regret or something of that sort. One difficulty arises because we don't have a unique perception of ourselves. Under different scenarios we think we will act differently- indeed, we will act as if we were different people. Thus one's own perception of utility involves comparisons between what satisfaction or regret we would feel under a particular scenario. This does not mean we have lots of different identities. It's just that we make choices which are consistent with the person we imagine ourselves as being in the scenario that has salience at the moment we make the choice. Obviously, being drunk or having been lied to, may cause that scenario to have little correspondence with reality. Still, by keeping in this mind we understand the behaviour of others- e.g. 'why did Tom propose to Sally? He was drunk. Also Mary assured him that Sally wasn't a lesbian and, what's more, she was the heir to a fortune.'
Since it is useful to have a theory about why people do what they do- and since this also changes what we do because we imitate the smart people while avoiding choices habitually made by losers- there is no harm in having a Utility theory which is based on things like maximizing usefulness or minimizing regret. The thing won't be 'categorical' because the two functions aren't dual. Still, for any restricted purpose, we can arbitrarily put in some numbers and get a useful enough result.
Turning to a theory mathematicians have found very useful- viz Von Neumann & Morgenstern utility- Ken Binmore explains it origins thus-
Modern utility theory famously began when Oskar Morgenstern turned up at Von Neumann’s house in Princeton one day in the early forties complaining that they didn’t have a proper basis for the payoffs in the book on game theory they were writing together.
Sadly, they also didn't have a proper basis for determining the range of options. As with 'payoffs', options are arbitrary stipulations made for a particular purpose. They may be useful. They may not. Not too much can be read into them. There is no uniqueness or non-arbitrariness to the result. Still, the thing may be useful enough.
So Von Neumann invented a theory on the spot that measures how much a rational person wants something by the size of the risk he is willing to take to get it.
in other words, he ignored Knightian Uncertainty- i.e. the fact that we don't know all possible states of the world and thus their associated probabilities- in order to arbitrarily put in some columns and rows and figures. One could say this was what the guy 'Expected'. But the truth is it is what it is convenient for us to expect the agent to Expect so we can do a bit of math rather than read tea-leaves or examine the entrails of an animal slaughtered for that purpose.
Binmore gives a jaunty account of the procedure-
Pick two outcomes, W and L, that are respectively better and worse than any outcome that we need to discuss. (One can think of W and L as winning or losing everything that there is to be won or lost.) These outcomes will correspond to the boiling and freezing points used to calibrate a thermometer, in that the utility scale to be constructed will assign 0 utils to L, and 100 utils to W. Suppose we now want to find David Hume’s Von Neumann and Morgenstern utility for scratching his finger (so as to avoid the destruction (to prevent the destruction of the universe). For this purpose, consider a bunch of (free) lottery tickets in which the prizes are either W or L. As we offer Hume lottery tickets with higher and higher probabilities of getting W as an alternative to scratching his finger, he will eventually switch from saying no to saying yes.
Because that is the assumption of the theory
If the probability of W on the lottery ticket that makes him switch is 73%, then Von Neumann and Morgenstern’s theory says that scratching his thumb should count as being worth 73 utils to Hume. Each extra percentage point added to the indifference probability therefore corresponds to one extra util. As with measuring temperature, it will be obvious that we are free to choose the zero and the unit on the utility scale we construct however we like. We could, for example, have assigned 32 utils to L, and 212 utils to W. One then finds how many utils a scratched finger is worth on this new scale in the same way that one converts degrees Celsius into degrees Fahreneheit. So a scratched thumb worth 73 utils on the old scale is worth 163.4 utils on the new scale. My guess is that Bentham would have been delighted with the mechanical nature of Von Neumann and Morgenstern’s theory, which reduces evaluations of individual welfare to ticking off boxes on a simple Gradgrindian questionnaire, but he would also probably have made the same mistake as many economists in over-estimating the extent to which real people make their choices in a consistent manner.
Real economists are very inconsistent in that they admit there is Knightian Uncertainty in the real world while also suggesting that their Expected Utility Maximization models can be the basis of 'Rational Expectations'. They should be supporting some type of regret-minimization theory.
Although the utility theory of Von Neumann and Morgenstern performs at least as well as any comparable alternative proposed by the new school of behavioral economists, it is only the rather poor best of a bad lot when it comes to predicting the behaviour of laboratory subjects.
Unless you cheat and tell them what they are supposed to do. Also beat them severely if they disobey. Sadly, even if you cheat, there will still be a 'duplication' problem.
Critics sometimes complain that a person’s attitude to taking risks is only relevant when gambling,
life is a gamble. You don't know whether you will get to eat the burger you just paid for because some crazy terrorist decides to stab you.
and so Von Neumann’s theory is irrelevant to the kind of moral situations to which utilitarians apply their ideas, but it is hard to think of a more fundamental moral question than: who should bear what risk?
Guys who know more about it or industries which have the clout to reduce it. Thus, the arbitrageur who knows the metal market should take the risk that copper might suddenly fall in supply while the Fire Insurance Company- which can persuade the Municipality to install fire hydrants- takes on the risk that your house will burn down.
Utilitarians who want to use the kind of insurance argument employed by Edgeworth or Harsanyi in defence of their position certainly have no choice but to accept that attitudes
to risk are basic to their approach.
Not Indian utilitarians. If you put all the burden on the rich/smart, they fuck off to a place where their taxes will buy them better public services- including Social Insurance.
Nor is there any lack of support from traditional sources. As it says in the Book of Proverbs: it is the lot that causeth contentions to cease.
Only if people agree to it and there is a greater benefit from getting on with the game than deciding who gets to bat first. Otherwise, if night is coming on and my side loses the toss I pick up my stumps and go home. This may cause a lot of contention but better a punch-up before nightfall then no fun at all.
Intrapersonal comparison. Why does a rich man hail a taxicab when it rains while a poor man gets wet?
Rich people take taxis. Richer people have chauffeur driven limousines. This is because they have lots of money and don't care who knows it. At least that used to be true of the better parts of London. Nowadays, my wealthy Indian friends are afraid to hail a cab lest their Rolex gleams in the sun-light and some mugger on a moped slices off their hand with a machete. Also, cabs don't stop for me in the rain because, judging by my clothing, they assume I am too poor to pay the fare.
Economists answer this traditional question by making an intrapersonal comparison. They argue that an extra dollar in Adam’s pocket would be worth more to him if he were poor than it would be if he were rich.
Adam may hail the cab though he is poor because he knows he has a weak chest. If he gets wet, he might get pneumonia and thus not be able to earn to provide bread for his numerous offspring. This is 'regret minimization'. It also explains why a rich man, afraid of being mugged for the Rolex he forgot tot take off, might use the excuse of the rain to run quickly to the safety of his 5 Star hotel.
But how are we to measure such increments in well-being?
There may be none. What is important is not doing something you'd later regret. At one time, I took black cabs day or night. A female friend of mine only used a particular female-driver minicab company at night. This is because a friend of hers was raped in the back of a black cab. Hearing this story, I decided to avoid black cabs with female drivers lest they rob me off the pearl of my virginity.
The unit of measurement certainly cannot be the dollar
sure it can. The thing is called 'consumer surplus'. The billionaire would have paid up to ten thousand dollars not to get wet in the rain.
because, for most of us, an extra dollar becomes less valuable the more of them we have—a phenomenon that economists refer to as the decreasing marginal utility of money.
or the increasing disutility of not spending it, if you have oodles of the stuff. But spending money isn't the same as giving it away.
The answer offered by the Von Neumann and Morgenstern theory is that one can measure the well-being of a person with decreasing (or increasing) marginal utility of money by counting the number of utils by which his total utility is increased when he gains an extra dollar.
Nonsense! They would just make up a number and stick it in. It is useful to speak in terms of money you would be prepared to pay because this gives us a notion of elasticity which is very useful when it comes to deciding what to tax. If there is a high opportunity cost (or 'dis-utility') there will be high elasticity. Don't raise taxes on it. Revenue will fall. You will have less to spend on the Poor.
Because a rational decision-maker in the Von Neumann and Morgenstern theory acts as though maximizing expected or average utility, he behaves as though each util is “worth” the same as any other.
Sadly, there can't be 'diminishing marginal utility for money' because Speculative Demand exists. The rich guy may borrow from his poor relatives to buy consols because he knows interest rates are going to fall. What one can do with money changes as you get richer. One option not available to those on Welfare is to emigrate to a country which isn't going down the toilet.
The rationality assumptions built into Von Neumann’s theory simply require
that people make decisions in a consistent way,
i.e. the fool will never learn from his folly. There are no mimetic or learning effects.
but his conclusions are surprisingly strong. Anyone who chooses consistently in risky situations will look to an observer as though he or she were trying to maximize the expected value of something.
Only if that is what they are determined to see. The fact is consistent choice may be required of us for some deontological reason. We are obeying a rule. We aren't maximizing or minimizing anything. This is also true of the behaviour of a guy who gambles because he likes gambling.
a Presbyterian minister is considering how to value his house or car. He will not regard the possibility that his house might burn down or his car be stolen as a possible source of excitement. He will make a sober assessment of the probabilities and of the value he places on his property, and then take out insurance to cover himself against the objectively determined risks he faces
Nonsense! He will call a few brokers and get a bunch of quotes. The actuarial models used to arrive at the quotes may be cast in a Von Neumann framework. But the Insurance Company as a whole needs to take an Enterprise wide approach to risk. If it doesn't, it may not be able to pay out. Still, hopefully some are too big to fail and thus taking a policy with one of them is 'regret minimizing'. Even if the worst happens, you won't reproach yourself too much. Plenty of smarter, more prudent, people are in the same boat.
As an example of the power of the theory, imagine that some bullets are loaded into a revolver with six chambers. The cylinder is then spun and the gun pointed at your head.
If you think you can buy the guy off, do so by all means. If you think he is a psycho, try to get in his head. Say 'This was foretold to me by this strange guy I met. He told me you would kill me but that he would avenge me. I didn't believe him. It's my own fault, I'm going to die today. The irony is, my rich Uncle is bound to drop dead any day now. But, it is my destiny to die a poor man.'.
Would you now be prepared to pay more to get one bullet removed when only one bullet was loaded or when four bullets were loaded?
Fuck does it matter? All you can pray for is that this nutter doesn't take his time killing you.
Usually, people say they would pay more in the first case because they would then be buying their lives for certain.
A guy who only puts one bullet in the chamber may be a sporting sort of chap. Pay him enough and he may let you go. Four bullets in the gun suggests a fucking maniac. Just pray your end is quick.
But the Von Neumann and Morgenstern theory says that you should pay more in the second case, provided only that you prefer life to death and more money to less. What conclusion should be drawn from such a conflict between one’s gut feelings and the recommendations of Von Neumann and Morgenstern’s theory?
It's just a theory. It may be useful to actuaries or financial analysts or whatever. If so, good luck to it. One place this theory isn't useful is in 'incomplete contracts' which is what the Social Contract is. This is because Knightian Uncertainty lies at the heart of it. The thing may look like a contract of adhesion and it may be so for immobile factors. But, long run, factors are immobile. 'Strains of commitment' weaken by 'common knowledge' if there is a lot of negative Voice and Exit. Loyalty may just be virtue signalling.
people are likely to be in a much more reflective mood when considering social contract issues. As the old form of English marriage service used to say, this is an occasion for making decisions “advisedly and soberly”
you have to consider whether she thinks you think she thinks you think she is a slut. If you do, your 'strains of commitment' will be weaker. You need her to think you think she thinks you think she will be faithful.
Fuck all that! She is like you. She has her doubts. You have your doubts. The whole point about getting hitched is you can both let them cancel each other out. Both of you are taking a risk but that also means you are in the same boat.
Marriage, in most cultures, is a 'costly signal'. It supports a 'separating equilibrium'. It is also a 'discoordination' game when compared to the usual coordination game associated with human sexual reproduction or, more realistically, passing the time coz there's nothing good on the telly box.
I think that the empathetic preferences with which we find ourselves holding are a product of social evolution.
We may or may not have empathy. We don't have empathetic preferences. We say 'Binmore is over six feet tall'. We don't say Binmore's height preference is in the above six foot range.
We need such empathetic preferences only because they serve as inputs to the equilibrium selection criteria that lead us to speak of “fairness” when we try to explain to ourselves what we are doing when we use them.
We don't need 'equilibrium selection criteria' when we speak of fairness. On the other hand, when we try to explain to ourselves how come Binmore is taller than us, we do need to input his height preference into some complicated equation.
However, it is important not to allow oneself to be deceived by this propaganda. Our “fairness” criteria do not necessarily treat all citizens in an even-handed mannner— whatever this might mean.
You might say 'it is unfair that Enzo Maresca has been dropped by Chelsea Football Club'. I point out that I wasn't permitted to play for Chelsea against Man U even though Mummy had given me plenty of sarnies and orange squash to hand round at half time. What's more unfair- a 45 year old being sacked as football manager, or an obese 62 year old being denied the opportunity to get a bit of physical exercise?
As with all social institutions, the “fairness” criteria we use will tend to result in certain types of behavior becoming perceived as more successful than others.
e.g. being good at football becoming a criteria for playing for Chelsea
Those whose behavior is perceived to be successful are more likely to serve as the locus for meme replication than those who are perceived as failures.
This is Tardean mimetics. We imitate those who are superior to us. That's why my twerking videos on TikTok haven't gained traction. People think Beyonce is better when it comes to shaking her money-maker.
The point here is that social evolution will tend to favor the survival of whatever empathetic preferences promote the social success of those that hold them at the expense of those that do not.
in other words, people who fake empathy will prevail. There's a cheap talk pooling equilibrium- everybody saying 'how are you?' and nobody actually caring about how you feel after being told you can't play for Chelsea even if you brought plenty of sarnies and orange squash with you to hand round at half-time.
In the medium run, an equilibrium in empathetic preferences will be achieved. In my books, I argue that, in such an empathetic equilibrium, everybody will have the same empathetic preferences and hence we will all share a common standard for making interpersonal comparisons of utility—as Harsanyi and Rawls suggest is actually true for our society.
We do have such a thing. It is called money. But it doesn't represent utility. It represents disutility or opportunity cost or transfer earnings. Thus Raheem Sterling gets 16 million quid a year whereas I don't get a penny from Chelsea. This is because Sterling can go to another team for about the same remuneration. His 'opportunity cost' is high. Mine isn't.
To clarify, consider the following scenario. You are offered a job, at one million quid a week, which involves sleeping with super-models. Initially you are very happy. You have high utility. Then you hear I get two million quid a week for the same job. Suddenly, you feel very hard done by. You experience dis-utility. You go through the motions with a succession of gorgeous women in a sullen and dispirited manner.
What is true of our society is that disutility matters because people can change jobs or alter the way in which their resources can be used. Those who have high elasticity of supply or demand have high opportunity cost or transfer earnings. Information about this could be encoded in 'Shapley values' or have some other such representations- but only in an ad hoc, ideographic or context dependent manner.
No comments:
Post a Comment