Thursday 2 August 2018

Sen's rich poet & why cardinal utility must exist

Is it possible to quote Amartya Sen in support of even the most plausible argument without immediately showing why it must be wholly and ludicrously false?

Consider the following excerpt from an essay in the Aeon magazine. The author wishes to present an argument against cardinal utility. It so happens this argument can be proven false for a complex mathematical reason. However, the moment Sen sticks his oar in, it becomes blindingly obvious that the argument must be false without any prior mathematical knowledge.

Sen describes a man bumping into a friend he hasn’t seen in years. Waving goodbye from his chauffeur-driven Rolls-Royce, the friend looks shockingly prosperous and well-heeled. Later, visiting that friend at his mansion in Chelsea, the man remarks upon what a high standard of living he observes. ‘Not at all,’ the old friend replies. ‘My standard of living is very low. I am a very unhappy man … I write poems, damn good ones too, but nobody likes my poems, not even my wife. I am always depressed about this injustice, and also sorry that the world has such deplorable taste. I am miserable and have a very low standard of living.’ The man has no reason to doubt his old friend, but feels obliged to remark he seems confused about the meaning of ‘standard of living’. To which the friend replies, ‘My standard of living high/What a plebeian lie!’, adding to the set of people who don’t think much of his poems.
A moment's thought shows that this rich poetaster is simply the victim of his own stupidity. He needs to incentivize appreciation of his poetry. Indeed, he needs to set up a Newcombe type situation whereby pedants will sincerely believe his poetry is the acme of aesthetic perfection for some hermeneutic reason they themselves will discover.

We do not say a rich man who starves to death because of some mental impairment which prevents him from ordering a pizza had a low standard of living. We say he is mentally ill.

Sen need not have given such a stupid example. He could just as easily have had his rich man say 'my life is impoverished by not being able to write good poetry.' This is perfectly reasonable. However it has no relevance to the notion of 'standard of living' which is statistical and applies to a broad class of people. Thus some other rich man, who does write good poetry, may feel his life is impoverished because he can't dance like Nureyev. These things cancel each other out. To brood over such deficiencies is a sign you need medical help- which is cool if you can afford it. The notion of 'standard of living' has nothing at all to do with such private obsessions or ambitions. Rather it serves a useful purpose in the discourse of Political Economy.

A person who enters your house and says 'I observe you have a high standard of living' is either a foreigner with a poor command of English or else suffers from a mental impairment. By contrast, saying 'The British standard of living has fallen by x %  because of Brexit' is perfectly sensible- though it may not be true.

The author, a research student, writes-
In this instance, one could define standard of living as the aggregate of pleasures from economic wellbeing and poetry, which differ in quality and are cardinally incomparable.
The author is wrong. Nobody, even Sen, has ever defined standard of living in this way because it would be utterly foolish to do so. Two people may have the same standard of living but very different levels of utility or happiness because one is old, lonely and dying painfully of cancer while the other is young, healthy and virile.

The friend assigns more weight to pleasure from poetry while the man gives pleasure from economic wellbeing a higher value.
Nonsense! The rich guy is just being stupid. His friend should say 'if your weighting for being appreciated as a poet is 90 percent then devote 90 percent of your income to securing your objective.'
The rich guy, with the shrewdness characteristic of his class, replies, 'I want you to arrange it for me, in exchange for 1 per cent of my income which is still much more money than you make in your miserable job as Professor of Literature' and the other bloke has to agree. He then publishes a book about how the rich dude's shite is 'ironic' and 'post modern' and comparable to that of Mayakovsky, Celan and Patti Obaweyo Golem.  Since the rich dude's shite is truly shite, he will soon be very highly regarded.

The truth, however, is the rich bloke has an easily accessible satiation point for praise. Indeed, he probably derives- as I do- greater utility by being able to think of himself as an unrecognized genius.

Consequently, their disagreement over the quality of life. In this case, the friend – the one living the life – is the best judge because only he knows what matters to him and how much. There is no way to aggregate the pleasures in Bob’s life unless those doing the sum assign values that are just like Bob’s.
Why would they want to do anything so foolish unless an arbitrage opportunity were available? In this case, some professional poetry critic or publisher could use this information about the rich guy's preferences to come up with a proposal for a new magazine or chair for poetry which the rich guy funds in return for securing a great reputation as a poet. Clearly it is important to specify the rich guy's cardinal utility function correctly so as to extract the maximum feasible 'rent' extractable by reason of this informational advantage.
In trying to assess pleasures differing in quality, the evaluation of an individual’s well-being must be done by that individual alone, says the economist Eric Maskin at Harvard. There is no absolute formula for calculating the pleasures in aggregate because the value is subjective, based on individual judgment, experience, and taste.
There is 'expert cognition' which does this extremely well. There are plenty of Godmen in the author's own ancestral homeland who started off with nothing and who died worth billions because they knew how to make calculations of precisely this sort. They then broker deals between 'disciples' so that the guy who wants to be a famous poet pays for another guy to get elected as the savior of Secularism who in turns signs off on some bunch of terrorists running amok. Nobody who watches Bollywood movies- or, indeed, Netflix's Sacred Games- doesn't know this.

The mathematician Barry Mazur at Harvard concurs: treating qualitatively different pleasures as cardinally non-comparable, he says, has both descriptive as well as predictive advantages over the view that we can weigh the pleasures for a sum total of them all. A better calculus would be allowing the pleasures to coexist, each independently in the form of a vector. With this kind of maths. the weight we give to each pleasure will be more relevant and our life decisions more satisfying in the end.

WTF? Weighting means a vector can be collapsed into a scalar measure of cardinal utility. But then any mutli-dimensional measure can always be made consistently unidimensional 'after the fact'. This may not be effectively computable but is good enough as a focal point. In any case, Economics is concerned with ergodic processes subject to the Law of Large Numbers. Individual idiosyncrasies cancel each other out. A quite separate point has to do with why Mathematical Economists should not be allowed to influence Public Policy. They are stupid and ignorant in a manner precisely the opposite of the way the rest of us find it pays to be stupid and ignorant and switch to covertly watching porn on our smartphone the moment anyone starts droning on about Social Justice and Human Rights and other such hypocritical shite.

No comments: