Monday, September 25, 2017

A Free Marketeer, but with a 'but'.....



Yes, free market economics is a feedback system! If there's a non-linear 
coupling here then we have all the potential for chaos!

I've been talking to James Knight on his Facebook discussion group about economics. He has, in the past, called himself a "libertarian" and he writes for the Adam Smith institute. Nowadays, however, "libertarian" just isn't a good label - it is term which has become blighted by association with the anti-establishment right-wing market protectionists and hyper-patriots in the US. In this context "libertarian" is a term giving pretext and plausible cover to a movement that in the final analysis, I submit, would ultimately prove to be anti-free market; after all, the anti-establishment disaffection we find in the US is in part the outcome of global trade changes, changes which, ironically, are to be expected as the free market works out its global logic. The reaction of the hyper-patriots to this global situation is to feel the draw of a consolidated nationalism (sometimes bordering on fascism and race supremacy) and  this encourages trade protectionism. These hyper-patriots are also fertile ground for a paranoid conspiracy theorism which imagines established government to be the seat of manipulative Machiavellian players. Let us remember that Hitler himself was an excellent exploiter of inter-cultural distrust and fear and populated the dark crevices of government with demons from his imagination.

Extreme libertarianism is anarchistic in sentiment and this, ironically, echoes the Marxist concept of the government-less commune, an idealistic state of society thought to be achievable via the pro tempore arrangements of the so-called dictatorship of the proletariat.  But given the self-deceits of human nature I have little doubt that the uncompromising idealism of the extremes of left and right, extremes which seek to do away with established (democratic) government, hide an all too human logic which ultimately results in the setting up of dictatorships, dictatorships which are anything but free-market friendly: Economic-hands-off tends to go hand in glove with intellectual-hands-off. Dictatorships feel threatened by this because it entails the distributed ferment of ideas which necessarily drive markets. Dictators (who sometimes have mental health issues) have a tendency to view the society they seek to dominate as an extension of their personality. But free markets do not usually thrive under the idiosyncrasies of dictators.

Over ten years ago I did a series on this blog called Mathematical Politics. This series briefly considered the question of centralized control verses decentralized free market control. The series ran into ten posts: Viz:

http://quantumnonlinearity.blogspot.co.uk/2007/04/mathematical-politics-part-3.html
http://quantumnonlinearity.blogspot.co.uk/2007/04/mathematical-politics-part-3_18.html
http://quantumnonlinearity.blogspot.co.uk/2007/05/mathematical-politics-part-5.html
http://quantumnonlinearity.blogspot.co.uk/2007/05/mathematical-politics-part-6.html
http://quantumnonlinearity.blogspot.co.uk/2007/05/mathematical-politics-part-7.html
http://quantumnonlinearity.blogspot.co.uk/2007/06/mathematical-politics-part-8.html
http://quantumnonlinearity.blogspot.co.uk/2007/06/mathematical-politics-part-9.html
http://quantumnonlinearity.blogspot.co.uk/2007/07/mathematical-politics-part-10.html

After 10 parts I got bored with the subject, abandoned it and returned to the physical sciences which I much prefer. But my recent involvement with James' Facebook group has meant I've had to blow the dust off this series. The last part finished with this:


It is ironic that both laissez faire capitalists and Marxists have faith in the power of a kind of “emergence” to work its magic. Both believe that once certain antecedent conditions are realized we are then on the road to a quasi-social paradisr. For the laissez faire capitalist the essential precursor is a free economy. For the Marxist the overthrow of the owning classes is the required precursor that once achieved will allow all else to fall into place. There is a parallel here with the school of artificial intelligence which believes consciousness is just a matter of getting the formal structures of cognition right: Once you do this, it is claimed, regardless of the medium on which those formal structures are reified, conscious cognition will just “emerge”. Get the precursors right and the rest will just happen, and you needn’t even think about it; the thing you are looking for will just ‘emerge’.

Rubbish

In other words I was deeply suspicious of any "panacea" prescriptions from both the left and right. However, in spite of that I still thought of myself as basically a person who was enthusiastic about the free market, although with a 'but'. For example, I finished part 4 with this paragraph:

So, the argument goes, for the successful creation and distribution of wealth the centralised planning of a command economy is likely to be less efficient a decision making process than that afforded by the immense decisional power latent in populations of people who are competent in identifying and acting own their own needs and desires. In particular, technological innovation is very much bound up with the entrepreneurial spirit that amalgamates the skills of marketeers and innovators who spot profit opportunities that can be exploited by new technology. Hence, free market capitalism goes hand in hand with progress. Such activity seems well beyond the power of some unimaginative central planner. It has to be admitted that there is robustness in this argument; Centralised planners don’t have the motivation, the knowledge and the processing power of the immense distributed intelligence found in populations of freely choosing agents.

But there is always a but.....

Always acutely aware that one must never be dogmatic about one's views, views which are best submitted to one's doubts and criticism, I then went on to consider conditions where the free market philosophy might need a bit of qualification.

But with today's partisan politics, in part a reaction to the economic instabilities caused by market globalization, there has been an increasing trend for the debate to polarize between left and right. The upshot of this is that there is a tendency to put people in either one of two camps: "If you are not with us you are against us!".  Given this context I suppose I shouldn't have been surprised when James focused on my 'buts' rather than identify me as a fellow free market proponent even after months of discussion. As I've already said my acceptance of the market system necessarily, repeat necessarily, comes with a fair measure self-criticism, qualifications and 'buts' (an essential part of my epistemic method as far as I'm concerned). I think James saw this self criticism as a failure of faith and a failure to "convert".  Here's a sample of the response he gave me: 

When we meet I'm going to try to see that you get to grips with how much you may be underestimating just how much trade has done for humanity, and by equal measure how much you may be underestimating the harm done to humanity by all the things I regularly criticise

Markets work for all the people that partake in them, because markets are cooperation through and through.

Where people suffer is when they have barriers to enjoying the market benefits - such as through state meddling, corruption, cronyism, war, etc.

I think you're still learning about the power of markets, Tim - you're still in the early days.

I know he's far from perfect, but I'd recommend reading Matt Ridley's The Rational Optimist - it's a fabulous intro to the development of markets and all the ways in which they benefit humankind.

This conjures up the image of me still loitering in the foothills of economic enlightenment, failing to tread the straight and narrow road to riches and therefore in need of full conversion! My "conversion" to "libertarianism" will never happen of course, because that would be contrary to my epistemic sentiments. I do all I can to avoid partisan contentions so assimilating my intellect that they become part of my personality; that would be to succumb to a kind of intellectual hegemony and commit intellectual suicide. So, rather than allowing some intellectual position to assimilate me, I instead endeavor to assimilate itThis means that wherever possible I maintain a studied detachment towards a subject, thereby avoiding becoming an enslaved "fan" to anything less worthy than the very meaning of life. Therefore I could not argue for the free market as a polemicist as does James; that's where we differ I think. 

In thinking about James reference to still learning about the power of markets  and underestimating just how much trade has done for humanity we find a huge irony. Marx himself seemed to be fully aware of the power of the markets. In fact in my copy of the book "Marx on Economics" edited by Robert Freedman, Freedman quotes Marx as saying in 1848:

The bourgeoisie during its rule of scarce one hundred years has created more massive and more colossal productive forces than all of the proceding generations together. Subjection of nature's forces to man, machinery, applications of chemistry to industry and agriculture, steam navigation, railways, electric telegraphs, clearing of whole continents for cultivation, canalization of rivers, whole populations conjured out of the ground - what earlier century had even a presentiment that such productive forces slumbered in the lap of social labour?

And here we are today well over 150 years on with progress that makes even 1848 look like a primitive world! Freedman comments on Marx' statement as follows:

Written more than a century ago, this is certainly one of the most eloquent testimonials ever written to capitalist achievements. One looks in vain to modern communist theoreticians for similar generous recognition of capitalism's contemporary achievements. 

The power of the markets is old news and therefore not news. I certainly feel that I've long since moved on from this realisation. Like Marx you may understand the power of markets but that doesn't mean to say you don't have reservations and 'buts' about a market driven economy, as did Marx himself, of course.  But your particular 'but' may not be the same as someone else's 'but'. In fact it's all a bit like one's attitude to cars. Cars are a convenient, comfortable and fun way for getting from A to B....but....perhaps you are a new ager who detests promethium technology, especially cars and would prefer them all to be scrapped just as the extreme Marxists want to throw capitalism in the dustbin of history. Or perhaps you like cars but think you can improve their performance by tinkering around with their mechanics; trouble is, if you don't understand the mechanics you may well end up wrecking a cars performance envelope. Or perhaps you love the benefits of motorized travel but you drive around with reckless abandon and without consideration for other road users or the environment. 


***

The self-maintaining organised systems we are familiar with. e.g. biological organisms, societal systems and religious cults, are mixtures of both distributed information & control and centralized information & control (I&C for short). Distributed  I&C circuits* react to local conditions and control the local response whereas centralized I&C circuits react to global conditions and coordinate a global response. In the societal case good communications technology allows  humans to be not only fully conscious of the global picture but also capable of organizing centralized I&C systems that effect society globally. That's not to say governments (=centralised I&C) can't and do screw up the natural workings of the market, but the point is that centralised I&C is just as systemically deep seated to sophisticated human societies as is the distributed processing of the market system. In particular, the huge surpluses generated by sufficiently advanced market based economies create the pressing question of who is going to control that surplus and how it is going to be distributed. Given human nature it is no surprise to find that there is great potential for dissent at the way a free market system distributes wealth. In response to the potential for defection (see my prisoner's dilemma point below) the centralised I&C systems (i.e. governments) are set up to settle/enforce ownership rights and this can be done either democratically or autocratically. Hence, it is likely that modern commercial industrial societies will always be a mix of distributed and central information & control.

It is very ironic that today's libertarian warrior is paralleled in the modern eco-warrior; both tend to be anti-establishment, but the parallel doesn't end there. The eco-warrior will point to the all but incomprehensible complexity of the web of feedback circuits found in the eco-system and how vulnerable they are to human meddling, meddling which is likely to adversely effect their fragile equilibrium. (Some eco-warriors have raised this to the level of an all but sacred principle: Gaia.) The libertarian warrior will say something very similar about the "natural" market systems. "Don't try to influence that system because its beyond human understanding". Like the libertarian's vision of the ideal laissez faire market the eco-system has little or no centralised feed back systems engaged in global control (unless you believe in Gaia).  But today's  intellectually sophisticated and technologically advanced humanity has the power and intelligence to centrally effect if not control both market systems and eco-systems for good or bad. True, a little knowledge allied to a lot of power can be a very dangerous thing and it's no surprise that both libertarian and eco-warriors are nervous at the thought of centralised human meddling. It is ironic that the decentralised markets which are at the bottom of so much technological and intellectual development, not to mention the emergence of powerful plutocrats, has had the effect of enhancing the ability of humanity to indulge in centralized cybernetic control (= centralised I&C)


***

As an aside let me comment that I would not trust Matt Ridley's general ideas in this connection.  In trying to use evolution to argue for doing away with the central planning (i.e. centralised I&C) Ridley shows no awareness that for conventional evolution to work it must start out with a huge burden of pre-ordained information. i.e it is effectively guided i.e planned in advance from an information point if view (See here, here and here). Organised biological structures do not simply emerge from an abyssal deep of pure randomness. Take a look at this synopsis of Ridley's book, The Evolution of Everything,  which appears on his web site:


Human society evolves. Change in technology, language, mortality and society is incremental, inexorable, gradual and spontaneous. It follows a narrative, going from one stage to the next; it creeps rather than jumps; it has its own spontaneous momentum rather than being driven from outside; it has no goal or end in mind; and it largely happens by trial and error – a version of natural selection. Much of the human world is the result of human action, but not of human design; it emerges from the interactions of millions, not from the plans of a few.

Yes, it may be a version of "natural selection" but if so then that means, as with conventional evolution, the "central" information constraining the system is implicit in the dynamics of life. True we may not (yet) understand where the pre-ordained information which drives natural history comes from but its presence challenges Ridley's notions about spontaneity, godlessness goalessness and not being driven from outside. Ironically Ridley's reference to "trial and error" actually gives the game away; generating trials can be modeled as a meaningless imperative process, but what determines the difference between error and success?  The concept of "success" has to be "centralized" and enshrined somewhere. Therefore, I propose, reality is a mix of both the imperative and the declarative. 

But having said all that I have no idea exactly what mix of decentralized and centralised management is optimum for a commercial economy (I'll leave the details of that for James to sort out!). But it may well be there are no pat answers to this question and that we should recall the lessons of John Holland's thesis about complex adaptive systems: That is, when chaos rules there are no equations or catch-all principles allowing us to construct definitive mathematical models enabling us to make definitive predictions or decisions which can be made far in advance. Consequently, a complex adaptive system such as a biological organism or a society copes with and adapts to a continuous stream of novel circumstances using trial and error feedback (i.e. Information and control). In such circumstances it is not possible to justify, using any general theory, a preconceived position which adopts in advance a particular mix of laissez faire (i.e. distributed I&C) and central I&C. In the complex adaptive system scenario adaptive behavior must be adopted because we don't know and never will know if a blanket "hands-off-the-market" strategy is generally justified; we just don't have any definitive mathematics to prove it. All we can do is adapt as best we can to the latest feedback from the environment. So, although I like to think of myself as an enthusiast of free market capitalism and its ability to generate innovation and wealth, I would never want to raise that enthusiasm to the level of a principle as some libertarian warriors seem to have done. The bug-bear with having to deal with polarised partisan polemics is that even though one may support a similar position, a bit of self-criticism and qualification comes over as putting one in the devil's camp.

Further to my many 'buts' I published the following list of issues on James Knight's discussion Facebook page urging him to take cognizance of these questions. This list has the potential to grow in the number of entries and the details of each entry; in fact I've already expanded it a bit since its first publication.

***

ONE: Do markets contain non-linear feed back systems? If so then chaotic market instability is a likely outcome. This could have an adverse effects (e.g. boom and bust) on working class people whose interests, in the main, go no further than wanting to live a hassle free comfortable life.  Given the limited quantum of the human life span these people may not be greatly motivated by the latest market generated efficiencies, of which they may never see the long term benefits and which to them only feel like unpleasant economic instabilities. (cf the great estates vs the swing rioters). 

TWO:  I put it to James that it may well be that free markets have a tendency to distribute wealth according to a power law. He seems to have picked this one up. However, there remains the question of whether this is a fact rather than just a conjecture of mine.  If it is a fact then the mechanism needs elucidating.  The power law distribution question is very relevant to the next question....

THREE: Re. the last two points: Prisoners dilemma and defection theory in the face of wealth inequalities, instability, short human lives and plutocracy. Marxism as an example of defection I suggest. It is often plausibly argued by left wingers that (power law?) wealth inequalities in capitalism creates a rich, powerful and undemocratic class of plutocrats, thus demanding centralized government in order to provide (hopefully!) a democratic forum to counteract plutocracy. 

FOUR: Limited ability of the market to solve certain classes of computation, particularly long term issues which don’t provide the incentive of local "internal" “pain” or “pleasure” signals. (i.e. “Externalities” is the jargon here). In particular I'm thinking of the ecological effects of human industry and wealth. But also, more abstractedly, the optimum market/government mix cannot be selected by the market itself; social systems are not sold over the counter and therefore are not subject to market selection. 

FIVE: Re the previous point. What information processor processes information external to markets. How can this information be passed onto the consumer as incentives to guide buying?

SIX: Market dynamics can be (crudely) understood and even (crudely) simulated (unlike politics). These understandings are, naturally enough, very tempting to  market interventionists, especially in the face of power law inequalities and the thrashings of chaotic instability. i.e. Markets prompt a concomitant centralised political response and therefore markets and politics will likely go hand in hand whether we like it or not. There may be a connection here with the correlation of a society's wealth  surplus and the emergence of government (cf Iron age Britain). It is an irony that compared to politics economics is relatively comprehensible - crude simulations of the market may be on a level with our understanding of atmospheric dynamics i.e. the weather. (However, probable chaos in both market and weather systems will compromise precise predictions and understanding).  All this is contrary to the libertarian idea that you don't touch markets because you don't understand them; the trouble is we do have a  modicum of understanding of market dynamics and that very understanding tempts interference!

SEVEN: Big production surpluses and increases in the overall wealth of a society lead to questions about the control and possession of that wealth, especially given human status driven motivations and the potential for human corruption.

EIGHT: Closely related to the latter point is this: Human beings are goal seeking systems (“complex adaptive systems”) and therefore will have a tendency respond in an adaptive way to their environment with declarative (that is "goal driven") social involvement. Ergo, modelling society as a pure “unplanned” imperative processes is unrealistic. Society has both declarative and imperatives influences impinging upon it. This point and the previous point aren't necessarily arguing for market intervention, but rather are pointing out the inevitability of intervention given that most real complex self-maintaining systems have available to them some kind of "central nervous system". 

NINE: Motivational theory: Commodities as a form of status symbol. The need for social status and a sense of community, belonging, purpose and identification. How well do monetarily valued market exchanges go toward satiating these all consuming human motivations and express human status choices? Is it likely that the human psyche can ever be completely satisfied with the way a free market distributes status values and community connection?

TEN: The “market of ideas": Perhaps not really a market because exchange doesn’t take place – ideas don't move about like conserved material commodities but can be copied and plagiarized with little cost. However, there does seem to be some kind of Darwinian struggle between ideas: Human beings are computational processors with limited computational resources and therefore can only entertain a limited number of concepts. The whole process looks to be more Darwinian than Smithian. Given that society is now linked by long communication links, this looks to be an important dynamic superimposed on society.

ELEVEN: Lastly the following point needs repeating (See ONE above): Are the feed back couplings of the market (see picture at the start of this post) nonlinear? If so chaos and market thrashings are a likely outcome when the market is subject to random perturbations. Under these circumstance the overall increase in wealth, measured as an average,doesn't carry a great deal of information if there is a huge background noise of troughs and peaks dwarfing the little human lives which live on these slopes. For some capitalism will feel like a bed of nails. 

This list has plenty of potential for expansion in length and detail. None of this necessarily attacks Smith’s minimalist idea that markets, if left to do their stuff, do a good job of increasing total wealth and should remain the core of any wealthy society. 



***

Finally I thought I would mention the libertarian paradox which I put to James and worded thus:

They (libertarians) claim the market should work for itself as government can't understand its inner workings (which is probably true) and yet how do libertarians know the market will work if no human mind can understand it?

James response can be found hereLet me quote part of that response:

But Tim's enquiry regarding how libertarians can know the market will work if no human mind can understand it doesn't strike me as being much of a conundrum, because in being asked to understand how the market works we are only being asked to understand that the market is society's aggregation of individual decisions by buyers and sellers made by people for whom those decisions brought about a mutual benefit.  

That is to say, while Tim is quite right that the free market is too vast and complex for politicians to understand its inner workings, it doesn't follow that because of this libertarians are on dodgy grounds assuming the validity of their position, because all us pro-market people are trying to say is that a free market in action must, by definition, be a system working for its agents, because it is quite simply the accumulation of activities that work for those agents.

There may well be unpleasant things in society that result from free market transactions, and concomitant power laws that cause discomfiture to certain socio-cultural groups, but us pro-market people are not primarily selling the qualities of the market per se - we are trying to advocate the freedoms from which things like the market operate more fruitfully.

James provides a very general answer here. If I'm reading him right then according to this answer all we need understand about the free market is that it is an expression of the bulk of consumer decisions and that each of these decisions entails an exchange of mutual benefit: e.g. one party wants and benefits from a competitively priced pen and the other wants and benefits from the money paid for it; what can be wrong with that? How can one be so crass as to deny that something mutually worthwhile has taken place and the market ensures that this happens millions of times daily? There is clearly an immediate local satisfaction here almost as if the whole population was engaged in mutual masturbation and getting immediate kicks. It can't be denied that accumulating the linear sum of the parts here it follows that there is a lot of pleasure/benefit entailed by market exchanges. However, as we know interacting parts are often more (and sometimes less!) than the sum of their parts. When a low level phenomenon such as local exchange of goods is aggregated there is the well-known phenomenon of emergence; that is, consequences at the high level which are difficult to anticipate without constructing a sophisticated simulating model. It is this difficulty in simulating an emergent outcome which I take to be the real content behind the view that we can't comprehend the emergent outcome myriad market interactions. However, James does acknowledge in the last paragraph of my quote those conjectured power law inequalities which I keep harping on about and the question mark which hangs over them, so we can't be far apart.

My answer to James blog post is this: Whilst at the low level, yes, markets generate a big linear sum of benefits, nevertheless that still leaves us with the question of emergent phenomena: If the claim is that we don't understand the complexity of all those myriad market choices then how do we know that it doesn't generate some unpleasant emergent effects? (Like dictatorial Marxism!) If you are spending all your days mutually masturbating one another, are you going to have your eye on the ball when it comes to things such as the ecological effects of human productivity, population explosion, the plutocrat in charge of a large media outlet or even nuclear bombs, social fragmentation and alienation, widespread Marxist disaffection cued by huge status inequalities, chaotic market instabilities causing unrest, technological advances perturbing the market and disrupting short lives, raw material shortages that take time for market driven technological solutions to fix etc, etc. Sometimes I have seen libertarian attempts to employ convoluted logical contingencies explaining why the free market will solve, say, the pollution problem after all. But this argumentation is usually carried out on a case by case basis with logic specifically tailored to the case in question and, unsurprisingly, always falling out in favour of the free market as a panacea. In physics when the same observation crops up again and again (such as the conservation laws) in widely different connections we don't expect it to be the outcome of a patchwork of miscellaneous and idiosyncratic proofs where the logic of each proof is entirely contingent upon the case in question; rather we suspect some general logic to be behind the conservation law in each case, logic which will one day, we hope, become clear.  So in the absence of demonstrably universal/general logic which might explain why free market economics is a catch-all solution, the idiosyncratic contingent logic of a case by case advocacy of the free market gives rise to the suspicion that a preconceived polemic is in fact the underlying and covert factor at work rather than some general principle.


***

In many ways capitalism with its restless adventurousness, sense of purpose, opportunities for creative innovation, its incentives, its concept of personal property & responsibility, production of wealth etc fits human psychology. But it doesn't fit it like a glove. Humans are not productivity termites and, moreover, the downsides of capitalism can lead to defections sometimes expressed as the cloud-cuckoo-land philosophy of Marxism which is a charter for dictatorship. It is likely that if elected to government Jeremy Corbyn's Labour party, although strongly socialist, would necessarily have to compromise with capitalism. But Corbyn's labour policies are very likely to depress the current industrial output leading to further dissatisfaction and the inevitable calls from the extreme left to go the whole Marxist hog and dismantle the UK's democratic and commercially based social order. The cry would, of course, be that you can't compromise with capitalism and therefore it and its state institutions must be cleared away for pure unadulterated Marxist communism. The extreme libertarian right argue similarly for the status quo version of capitalism; it's not working well, they claim, because we haven't gone the whole hog with unadulterated free marketism. The idealists of both  extremes seek social nirvana and don't tolerate the muddle, contention, compromise, and half way house solutions that are the natural state of human affairs. The other thing which the extremes of left and right aren't too partial to is system theory; that's just a complication clouding the cartoonish clarity of their thought.


Footnote
* "Information & control" is another name for "cybernetic circuits"


End-note on planning
Clearly large commercial projects like the development of the latest car or the building of a cruise ship involve huge levels of centralised I&C. Moreover, a large commercial conglomerate also entails a high level of centralized I&C. However, there comes a point when human cognitive, epistemic and organisational limits stultify the ability to influence outcomes via centralized I&C; for example the industrial revolution was obviously not an outcome of central I&C, but  was an emergent phenomenon arising from the net effect of much more fragmentary I&C systems. Moreover, technological innovation is a function of nature and of course we can't predict what nature will throw at us via the next scientific discovery; this defies the foresight of centralized I&C. However, within  human  cognitive and epistemic limits it is clear that central I&C is very much part of human nature and as information and processing technology improves centralised I&C is likely to be enhanced.  It is probably no coincidence that the first big cities appeared around the same time as writing appeared, the latter enhancing centralised information and control.