Tuesday 17 September 2019

BhuVai- a Tambram Chemist's Utopia

Bhuloka Vaikuntham means Paradise on Earth. Vaishnavites don't want Heaven or Gnosis. They want to remain on earth serving 'Daridra Narayan'- the God amongst us who is lonely and poor. Though, strictly speaking, they reject the notion of a jivanmukta (one who has achieved liberation in this life), yet the True Vaishnavs appear to have this quality. Though in the world, they are not of it. Though their cares and worries are the suffering of others; by taking on those cares and worries, they diminish that suffering. They represent 'Economia'- a type of management which is mutual, accommodative, merciful and joyous- as opposed to rigid, rule-enforcing, 'Akrebia'. These Greek concepts have a corollary in every Spiritual tradition. The Japanese 'peasant-sage' Ninomiya fused Confucianism, Shinto & Buddhism into a practice of agricultural economics which Christian Missionaries found inspiring. They translated a book about his work into many languages. A copy of this book is to be seen in Mahatma Gandhi's library- though it is not known if he read it. Perhaps he did and was repelled by the story of a Buddhist monk who takes up fishing to provide the poor with a source of protein. I mention this because 'Akrebia' when it comes to 'Ahimsa'- i.e. the Indian reluctance to violate the prohibition on taking even the life of a fish- constrained Hindu 'economia' from ever actually lifting the great mass of the people out of Malthusian poverty. Similarly, though Sufi savants like Fariduddin Attar had shown how Nishapur had been turned into a 'workfare' Utopia, Islamic jurisprudence reacted to the shock of the Mongol invasion by embracing beggary. Attar was saying no one should be a dervish. Everyone should be enabled to work and rise up. Rumi- who ran away to Anatolia (hence his sobriquet)- thought it was safer to be a dervish, preferably one who was completely insane, because otherwise you just end up being the slave of some genocidal Great Khan or the other.

In recent years, Hindus and sub-continental Muslims are increasingly influenced by our highly educated and successful diaspora in America. Unlike us, they have shaken off the miserabilist Akrebia involved in becoming a 'Sadhu' or a 'dervish'. Moreover, in contrast to British trained Indian intellectuals, these Americanized Indians aren't holier-than-thou elitists interested only in 'playing the race card' and leveraging their connection to a vast, seething, 'subaltern', mass of suffering, into personal advancement or sinecures in various Universities and NGOs and UN organizations.

As a case in point, l may mention Venkat Venkatasubramaniyam- whom I assume, on the basis of his sonorous name, to be a Tamil Brahman by ancestral origin- who is a Professor of Chemical Engineering concerned with the problem of economic inequality & distributive justice. He has coined the term 'BhuVai' to signify an ideal society whose members are equal in ability. Is there a way to also render them equal in happiness? Venkat believes so and he sets out his theory in a well-written, well-researched, book titled 'How much Inequality is Fair?'

In the Past, the basic problem for 'distributive justice' was thought to involve the theory of Value. The problem is that the Knightian Uncertainty of our fitness landscape means Value itself is Uncertain. Moreover, co-evolved processes defy the mathematics of complexity, getting rid of hysteresis but at the price of rendering ergodicity 'anything goes'.

 Breaking up ensembles by means of 'dis-coordination games'- spatially, this happens in 'Tiebout sorting'- permits mimetic, or bandwagon, effects to have a stabilizing, rather than runaway, effect. Incidentally, this would yield 'proportionate growth' in line with Gibrat's Law and hence generate a log-normal Income distribution. But, this does not mean such a distribution is 'fair' or 'natural'.

An alternative way of looking at this is to use the theory of congestion, or potential, games- indeed, this is what Venkat does. Here the Muth rational solution is to have an Aumann signaller dividing up the population. Thus, to reduce congestion, we may have a public signal- e.g. all those born in the first half of the year can use the congestible resource at such and such time. Those born in the second half of the year get a different time-slot. My point is that correlated equilibria based on public signals reduce the price of anarchy like nothing else and therefore we must expect them to arise and to 'partition' the relevant configuration space. This militates for a Schutzian 'ideal type' theory being heuristically useful or even approximating to the Muth rational solution.

In any case, from the point of view of Mechanism Design, there must, by the Spilrajn extension theorem, be a 'Revelation Principle' such that a Regret-minimizing, Hannan Consistent, multiplicative weight update algorithm exists which is 'natural'- i.e. similar to the sort of algorithms we believe Nature itself uses. This allows us to demarcate Nomos from Phusis. We can then create 'positional' goods to reflect the spread between the Imperative and the Alethic- i.e. Values and Facts- so this turns into a 'mechanism design' problem. However, doing nothing is an even better option. This is generally the case for any armchair exercise. Catatonics and Corpses are the most suitable candidates for conducting this type of research. At any rate, this is the common sense view. Nevertheless, there are elderly mathematical economists who have been discovered to be wholly useless and who are neither catatonic nor cadavers. A sense of 'there but for the grace of God goes I' militates for letting these pedagogues pretend to be Prophets of a new dispensation.

When I was 16, I attended the lectures of Professor Morishima (who had been chased out of Japan by his students) at the LSE. He thought he had solved this problem- thus reconciling Karl Marx and Adam Smith- with a 'rational distribution of Income' based on 'shadow prices'- i.e. the scarcer 'ability' gets a higher reward. The problem here is that 'Shapley values'- which factor in 'bargaining power'- are not aligned with 'shadow prices'. In other words, a Sociological 'Game theory' is at odds with the Production Engineer's view of the World. In 'Mechanism Design' (which is a reverse Game Theory) there is a trade off between incentive compatibility and efficiency. The theory of incomplete contracts is now the cutting edge of this type of research. However, from a historical point of view, this conflict simply recapitulates Pareto's own trajectory- he started off as an engineer (discovering power-laws in Income distribution in the process) in the laissez faire tradition before transitioning to Sociology with its sticky 'residues and derivations', thus becoming an intellectual pillar for everything from Mussolini's Fascism to post-Pigouvian Welfare Economics.

Interestingly, back in the 1950's in Venkat & my native India, two White savants, both Leftists, came up with what is known as the Goodwin 'Class Struggle' model- a predator-prey model in which Workers are the Wolves and Capitalists are the Sheep!

Turning to Venkat's 'Bhuvai' which is more in the Econophysics vein, it would be interesting to compare it to the recently published magnum opus of Anwar Shaikh (whose Mum was South Indian) and his Marxist analysis of 'turbulent flow dynamics' in Capitalism. Both Venkat and Anwar are good and decent men who write clearly and whose motivation is entirely good. However, in this branch of scholarship it is vain, worthless, hypocrites- people like Nussbaum, Piketty & Sen- who win the jackpot. Thus has it always been. Consider the notion of entropy and its possible application to economics. Venkat quotes Samuelson derisive comments on any such possibility. Yet Samuelson must have been aware that smart Jewish emigrants to the US- like smart Russian or German or even British emigrants- had used such notions to critique Capitalism. Their motivation was moral. In general, these guys worked hard and did quite well for themselves and for society. Anwar and Venkat have certainly done well by their students and made a very positive contribution to America. But, to hit the jackpot and advance themselves, they'd have been better off writing hypocritical, holier-than-thou, shite and playing the race card at every convenient opportunity. Thus, the reason their books on Econ aren't making them very rich is also the reason that Income and Wealth inequality has burgeoned. The good guys lack sociopathy. Yet, it seems to me, they populate an ontologically dysphoric 'BhuVai' which enriches our own world.

Venkat writes-
we study a competitive, dynamic free-market environment in BhuVai, comprising a large number of utility-maximizing rational agents as employees and profit- maximizing rational agents as corporations. Let us assume an ideal environment in which the free market is perfectly competitive, transaction costs are negligible, and no externalities are present. 
If no externalities are present, then there can be no coercive service industries- e.g. Police services or Armies. Also, there would be no corporations because there would be no Coasian need to 'internalize externalities'.  Thus, the solution concept is the folk-theorem, or Myerson general feasibility theorem, of repeated games. There may be Aumann 'signallers' permitting better correlated equilibria. However, Venkat proceeds to rule out this possibility by stipulating that there is no way to discriminate between signallers. Thus, either everything is 'common knowledge' or else there is no way to extract signal from noise.
There are no biases due to race, gender, religion, etc. We assume
that no single agent (or small group of agents), whether an em-
ployee or a company, can significantly affect the market dynamics;
i.e., there is no rent seeking or market power. In other words, our
ideal free market is a level playing field for all its participants.
However, under these circumstances, there would be no market. Society would be a wholly non-transactional repeated game- like family life. Mummy does not demand payment for nursing the baby. Daddy does demand money from Mummy for sexual services so he can go to the pub and get drunk, but Mummy beats him and makes him get a job as a Cost & Management Accountant instead.
We also assume that neither the companies nor the employees engage
in illegal practices such as fraud, collusion, and so on. We also as-
sume moderate scarcity of resources.
So, Venkat is describing a situation where a purely relationship based Society could thrive. History shows us that such Societies break down when there are exogenous shocks in demand or supply. That is why Venkat & me aren't living happily in some rural Agraharam.

Relationships are 'incomplete contracts'. The Market is a way of turning some 'incomplete contracts' into complete contracts of adhesion. However, there must be 'missing markets' unless complete fungibility- i.e. interchangability between commodities, but also agents- obtains. But, in that case, distributive justice is easily achieved by 'antidosis'- i.e. letting any agent swap places with any other agent. In ancient Greece, if your tribe decided you had to bear the expense of a liturgical duty, you could get out of it by offering to swap estates with some other guy in return for his discharging the duty. This is like the solution to the cake-cutting problem. Clearly, this is impractical. When Mummy beats Daddy for wanting to go to the pub, he says 'okay, let's switch places. You go be an Cost and Management Accountant. I'll stay home and breast-feed the baby. You just see, after a miserable day at the office, you too will need to go to the pub to relax and blow off steam.' Mummy beats Daddy some more because she knows nipples on a man are as useless as Amartya Sen as Chancellor of Nalanda University.
In our ideal free market, employees are free to switch jobs and move between companies in search of better utilities. Similarly, companies are free to fire and hire employees to maximize their profits. We also assume that a company needs to retain all its employees in order to survive in this competitive market environment. 
WTF? Why make such a foolish assumption?
Thus, a company will take whatever steps necessary, allowed by its constraints, to retain its employees. Similarly, all employees need a salary to survive, and they will do whatever is necessary, allowed by certain norms, to stay employed. 
Venkat has clearly never had to supervise an ordinary bloke like me. Elsewhere he says- 'The intrinsic properties of an employee are innate attributes that are exactly the same for all employees.'
If this is so, employees are 'fungible'. This means there can be an open market for them. But this means you buy their labor service on a just in time basis, you don't employ them. Thus there are no enterprises. Everybody buys and sells labor services on the open market in a frictionless manner.

Venkat's motivation is different. He thinks he is solving a deontological problem- 'These are the fundamental rights, such as the right to life and liberty, to a discrimination-free, healthy work environment, etc. 
Why employ anybody if you have to provide remedies for any breach of their right to life, liberty etc? You can just buy the labor service you need on a just-in-time basis.

These ought to be guaranteed to be absolutely the same for all the employees, and all employees are to be treated equally in this regard. This requirement in our theory corresponds to the first principle, the liberty principle, in the Rawlsian framework, where every individual in a civil society has equal basic 
liberties. Rawls arrives at this fundamental principle of equality through the application of his “veil of ignorance” concept to a group of rational agents in the original position. In our theory, we arrive at this by using the maximum entropy principle, which we discuss in chapter . 

The maximum entropy principle says ' the probability distribution which best represents the current state of knowledge is the one with largest entropy, in the context of precisely stated prior data.' Given the data Venkat has chosen for his gedanken, we know that maximum entropy is achieved when nobody is an employee. Labor services are fungible and sold on a frictionless open market. Everybody is his own employer and must safeguard himself from sexual harassment by himself- a full time job for wankers like me. Yet, I don't get paid to do it so I have gone on strike and thus am sexually harassing myself even as I type these words with one hand.

Venkat offers me some encouragement for my view that the Government- not of any existing State, but perhaps that of 'BhuVai'- will pay me for protecting myself from sexual harassment by me because it is only fair that I be adequately recompensed from my labor in this regard.

Venkat says-
It is important to emphasize that the free market itself, ideal or otherwise, is a human creation and “does not exist in the wilds beyond the reach of civilization”
Actually, the reverse is the case. Free markets only exist on terra nullis- i.e. 'in the wilds'. If there is a civilization, then there is a coercive contract enforcement mechanism. There is no 'free market'. All you have is a regulated market from which Manorial (or Tiebout) rents are drawn either by a 'Stationary Bandit' or by a Municipal Authority representing Civil Society.
(Reich 2015, 4). As Reich (2015, 3–5) observes, “Few ideas have more profoundly poisoned the minds of more people than the notion of a ‘free market’ existing somewhere in the universe, into which government ‘intrudes.’…A market—any market—requires that the government make and enforce the rules of the game.
Utter nonsense! Which government is enforcing 'the rules of the game' when it comes to drugs or prostitution or bribing senators?
In most modern democracies, such rules emanate from legislatures, administrative agencies, and courts. Government doesn’t ‘intrude’ on the ‘free market.’ It creates the market.” 
Sheer poppycock! Markets are created by Supply and Demand. Governments can't create a Supply of 18 year old virgins who want to sleep with fat, middle aged, very poor, men like me. On the other hand, if I had 'effective demand'- i.e. lots of money- I probably could find a girl ready to make this deeply repugnant sacrifice so as to rescue her family. This is why, I personally don't want a laissez faire utopia. Repugnancy markets- like the prostitution of innocent young people- would burgeon. I want cops to come and beat the shit out of pimps. I want a social welfare safety net, so young people can pursue their dreams. I also want good people, who have done useful work, to think about how to make 'BhuVai' a reality. Even if their general scheme is silly, they may have a practical 'mechanism design' type suggestion which improves at least one existing market or non-market mechanism.

Venkat, nice guy that he is, describes an ideal company which compensates workers according to their transfer earnings (which, in his model, would be their shadow price). He does not get that being required to simulate sociopathic behavior imposes a high psychic cost on non sociopaths. Most people don't want to be the guy who fires people. But, some people like firing people because they are sociopaths. Repugnant but necessary activities- e.g. firing the drunken Accountant who keeps bringing his baby to work and then tries to breast-feed it- get delegated to the sociopath who therefore commands a higher 'threat-point' independent of his shadow price. This is a story we have seen play out on a thousand TV shows and Movies since the time of J.R on Dallas or Gordon Gekko in Wall Street.

I recall my first job as a shop assistant. Suddenly the Assistant Manager- a weedy looking specimen- ordered us all to take off our shoes. He then came and searched us and looked at our socks. Why? He had found the discarded plastic wrapper of a pair of socks and assumed one of us had stolen the pair. Thus, for the theft of a measly pair of socks, this guy was keeping us in after work and treating us like criminals! I was fuming with anger at this humiliation. Some of my co-workers were tough guys who carried knives. I asked them whether they were going to teach this Assistant Manager a lesson. They averted their eyes and explained that his bonus was directly linked to 'stock shrinkage'. His anger was justified because money was coming out of his own pocket. He had 'skin in the game'.

Mechanism Design can get us to behave like sociopaths but sociopaths still have the advantage because they do the thing naturally and for pleasure. The same is true of risk-taking and telling lies and being a Nobel Prize winning Professor of Economics. Behavior we normally have to be compensated for, comes naturally to sociopaths. The smarter among them work out how to sell this ability of theirs to the highest bidder. Be it under Capitalism or Communism or Gandhian Socialism- the scum always rises to the top.

Venkat gives a brief account of Game theory and mentions that it can be found in the Talmud as Aumann has described. However, like most Indians, he is unaware that the first text to explicitly state that the 'Just King' (Dharma Raja) must study statistical game-theory to overcome his 'vishaada' (mental malaise involving choice theory) was the Mahabharata. Just as the 'Bhagvad Gita' helps Arjuna overcome his 'vishaada', so to does the Nalopokhyanam (along with the Vyadha Gita) help Yuddhishtra to overcome his fatal addiction to skill-less gambling.

Venkat may not know the work of the Jain polymath Umaswati. However, his ontology features an entropic end state or heat death wherein all beings attain kevalya. This concept was developed in parallel with Vaishnav notions of 'BhuVai' and is complementary to it. Umaswati, Nagarjuna and Sankara have different, but 'observationally equivalent', ontologies and epistemologies. Vaishnavs are not backwards in this respect. In other words, there is a synoptic Indian tradition in such matters and it can work very well in promoting the common-weal save when it runs up against the narrow Akrebia or deontological principles like 'Ahimsa'. But Indian jurisprudence has a workaround for this. The problem is that Hindu 'public intellectuals' are completely ignorant of Indian thought. They are credulous and naive when it comes to shitheads like Rawls and Dworkin. They also believe, against all the evidence, that Amartya Sen isn't a complete waste of space.

Is there anything valuable in Venkat's book? Yes. It is well written and sincere. Students should read it. I may write another post about some specific aspects of his model, but his quest is futile- its Grail is a crock of shit. On the other hand, if we regard it as what ethologists call 'displacement activity', then we should measure its worth by the other things Venkat and his colleagues are doing in their lives. I imagine these things would be very good and worthwhile in themselves. The same could be said of Yoga. The thing doesn't help others directly, but- if practiced by good people- it makes them better at serving others, which is their true path to 'Union'- which is what the word Yoga means.

I am not saying that there is no 'Revelation Principle' such that a Grothendieck Yoga of Mathematical Economics might not have univalent foundations. We know, a priori, that this must obtain and that is sufficient to create Schelling focal solutions which are 'indicative', not substantive or computable. That's good enough. A smart guy could write a long book describing this but the map would be larger than the territory. Why? Co-evolved processes get rid of complexity faster but less robustly than it can be generated. There is 'anti fragility' but the release of evolutionary capacitance depends on the future fitness landscape- so the underlying process is ergodic in one sense but not memory-less at all. Currently, the socio-technological landscape has become more uncertain. Thus, non-sociopathic Muth Rational agents, behind the veil of ignorance, would prefer a more skewed Income and Wealth Distribution because (1) they are only concerned with Friedman type 'Permanent Income' and (2) they are regret minimizing. What I mean is this. For an ordinary bloke like me, it is more difficult and depressing to suddenly become poor and lose my customary life style than to live high on the hog in good times and return to penury during a down-swing. So, it is regret minimizing for me to prefer that asset bubbles affect only sociopaths or 'lucky gene' legacy holders. Both may find an ordinary life more satisfying than their enjoyment of a tawdry type of wealth and privilege.

One final point. Venkat is a Professor of Chemistry. Yet his English is excellent. Why is it that our Professors of English Literature write like shit while our Chemists and Doctors write beautifully and clearly? The answer is that Higher Education in STEM subjects is not a Credentialised Ponzi Scheme. Mathematical Economics, sadly, can make no similar claim.


No comments: