Robots in demand in China as labour costs climb.
- This topic has 117 replies, 12 voices, and was last updated 6 years, 10 months ago by
.
-
Posts
-
What effect will automation on this scale have on Capitalism? Is it a problem or will it cancel itself out? Cheers Ray
http://www.guardian.co.uk/world/2012/nov/13/robots-demand-china-labour-costs
Our blog SOYMB has touched on this issue a couple of times. Herewhat it said. Foxconn has been heavily hit by labor issues over the past few years. To improve productivity and product quality it has declared it will introduce 1 million robots over the next three years to replace many of its workers – that it equals the number of operational robots worldwide.Once the cost of labor exceeds the cost of deploying robots, it is clear what a company will have as its workforce. Robots can work 24 hours a day, need no incentives to stay or payments to social security, but instead provide better output for routine work with higher efficiency and accuracy. Although the upfront investment is significant, established Chinese companies can afford it.Taking the molding industry as an example, the average salary of a worker is 1,500-1,800 yuan ($230-280; 180-220 euros) a month. Adding the cost for food and accommodation, the total cost for a worker is at least 2,000 yuan a month. As a robot can only execute certain tasks, 20 percent of the work remains to be done by a human worker. With the price for robots ranging between 20,000 and 30,000 yuan, the investment can be usually recovered within one to two years. The prices for robots have fallen as a result of technological improvements and the beginning of mass production. Robots have an average life span of 12-15 years, so the economics for a decision in favor of machines can be compelling.http://socialismoryourmoneyback.blogspot.com/2012/01/happy-chinese-new-year.html====================Foxconn already had 10,000 robots and would increase the number to 300,000 next year and a million within three. Wages in the region have risen by around a third over the past year, experts estimate, as the proportion of young workers shrinks and their expectations rise. Manufacturers are seeking to improve productivity, or shift production inland or overseas, as Foxconn has already done, with huge new plants in Chengdu, Chongqing and Zhengzhou and one site in Vietnam.[update, also in the Philippines and now rumours of a factory in the US]"As labour costs rise, companies will have to invest more in automated facilities. But we shouldn't get carried away; there is still a lot of cheap labour out there," said Tom Miller, of Beijing-based economic consultancy GK-Dragonomics.Alistair Thornton of IHS Global Insight, suggesting other companies would follow suit. "Workers can command higher wages and are less willing to settle for lower ones. You can no longer just double your workforce to double your output.""I am suspicious," said Liu Kaiming, of the Institute of Contemporary Observation, which supports workers in Guangdong. "Machines can do it, but think about the cost … overall, workers are still much cheaper. This is probably just for sensational effect, to put pressure on workers."Prof Huang Renmin, director of the institute of labour market research at the China Institute of Industrial Relations, agreed. "This is the trick capitalists use to threaten workers," he said.SOYMB suspects both views have truth. What improved machinery must do for the capitalist, if it is to justify its cost, is to increase the productivity of his workers (his rate of extraction of surplus value from them) to give a rate of profit equal to, or above, the average. The greater the outlay on fixed capital, however, the more dependent such profitability is upon the continuing buoyancy of the economy. But, of course, capitalism is incorrigibly cyclical and unpredictable. Expensive machinery lying idle in a slump represents growing losses for capital, to be set against the profits made in better times. In slumps, too, the real price of labour power is driven down—as we have seen in the first half of the 1980s. This makes more labour-intensive processes relatively cheaper and more profitable, offsetting the advantages of capital-intensive methods. Moreover, labour can be dismissed with a week's notice. Machinery can not. These considerations have made capital particularly slow to introduce automation, especially when wages are low. Capitalism is in a bind – it wants to use as much labour as it can as little as possible.http://socialismoryourmoneyback.blogspot.com/2011/08/chinese-capitalism.html
Production by Recursive Robotics — firms making robots making robots making consumer products competing against rival firms making robots making robots making competitive consumer goods.In this Ideal capitalist community, according to Marxian economic theory, the robots cost nothing [have no value], the consumer products are free [have no value] and the firms get zero profit. The capitalist day of judgement is at hand!For a real capitalist community, rushing headlong into this capitalist-production abyss, Marxian economic theory asserts [at least in a directly naive application of it]1. the organic composition of capital skyrockets2. the value of constant capital plunges3. the market price of [the recursive-robotically produced] consumer goods falls4. the social rate of profit plungesIf recursive-robotic manufacturing becomes the social norm, descent into the abyss must accelerate through increasingly desperate attempts to realize ever-diminishing rates of profit.Production by recursive robotics therefore looms as the great test for Marxian economics. It may very well be the great test for Marxism itself, and so for Socialism, and by implication for all mankind.Sure, we have seen wide-spread robotics before, and we have been alarmist before. But, …If the trend toward production by robotics becomes competitively recursive, increasingly minimising capitalism's dependence on a productive working class [variable capital], the implications for the world's working class and capitalist class, and for the existing capitalist and future socialist systems of society are enormous.
I covered this in a Standard Article (and talk) some years ago. One possibility is that price and value totally separate, and computer generated prices, from robot markets, come to predominate (the mechanism would be absolute rent, charged on intangible fixed assets), that and taxation may well slow down the collapsing rate of profit. The other, scarier prospect, is that capitalist growth grinds to a halt, and decadance ensues, with huge megaslums of surplus population growing even in the advanced countries. With no economic muscle, the working class would be prey to strong centralised states, administering state capitalism, as the last capitalists fight between themselves for the diminishing profits. I'm afraid robots can't replace class struggle politics.
Young Master Smeet,Thank you.Your interesting analysis nicely supports the case I was attempting to make in other threads that a capitalist social base has many actual possibilities. We can only understand these possibilities in terms of the base that supports them. If your analysis ever turns out to be actualized [as I said, my "analysis" was only a directly naive application of Marxian economic theory] it will demonstrate how important it is for the Party case to be as scientific as possible — exposing capitalist actuality as being deterministically consequential upon the base of capitalist ownership and control of social production.The non-socialist implications of recursive-robotic production are terrifying. They make the socialist case ever more urgent!
So what you are saying is that the working class could end up becoming a mass of Epsilons whilst the competing Alpha's control Frankenstein style dystopian states…with the prospect of class consciousness more and more unlikely without access to all of the essentials of education and life in general?Master smeet could you give me a link to the article you wrote? Aren't there some in the party who believe the problem of "Recursive Robotics" (a term I've never heard of) is a misnomer because the task of actually building and designing robots will make up the shortfall of other lost jobs which are irretrievably lost. I don't know enough about this issue. Does the party have one single stance on this?
Ozymandias wrote:Does the party have one single stance on this?I'm not sure about a single stance, but this classic article from 1965 by Comrade Hardy put exaggerated claims (either way) about automation into perspective. The same applies to robotics:http://www.worldsocialism.org/spgb/socialist-standard/1960s/1965/no-725-january-1965/automation-perspectiveComrade Hardy had already made the same point in an article dating from 1934:http://www.marxists.org/archive/hardcastle/production_productivity.htmAs he pointed out, productivity (which automation, computerisation, robotics, etc increases) only increases relatively slowly, at between 1 and 2% a year. This is because what needs to be taken into account is not the labour saved at the final stage of the production of some product but the labour saved on producing the product from start to finish. This is just Marxian economics and an application of the labour theory of value..It has also been borne out by the facts about what happened since 1934 and 1964. None of the dire threats or promises predicted at the time and mentioned in the articles materialised.Just read the claim made in 1934 that "that mechanical progress has gone so far in the direction of increasing output per head that we are within measurable distance of 'one immense robot factory, employing no workmen at all'". And the "forecast" made in 1964 that "'before the end of the century,' in every industrial country, certainly in the West, most of the essential work will be performed by about 20 per cent of the people – chiefly the most intelligent".I suspect that the same will be seen in 50 years about what some are saying today about robotics — unless of course socialism has been established by then.
You may find this link of interest.http://english.pravda.ru/science/tech/14-11-2012/122802-robots_food-0/Robots in farming
Four observations…Observation 1. Hardy is describing actual capitalist production as we know it to be.Current capitalist production can be described abstractly as Production by Workers — firms of workers using automation [or robots] to make consumer products competing against rival firms of workers using automation [or robots] to make competitive consumer goods.Hardy makes the point that automation reduces the variable capital [human working hours] in the "final" stage and not in the "preliminary" stages of the production of a commodity, and so the overall reduction in variable capital in the consumer commodity turns out to be much less than it actually is in the "final" stage of production alone.That is Hardy's reminder of Marx's representation of a commodity's value as being made up of two components [one living and one dead].1. Past Labour — the Means of Production that comprise the resources and depreciating instruments that are productively consumed in producing the commodity2. Present Labour — the "final" Labour that is productively consumed in producing the commodity.Hardy's point is predicated on only the "final" labour being displaced by automation and not the "preliminary" labour.It is hard to see this situation still characterizing capitalist production today, where all stages of the production process are increasingly automated and robot-ized.Observation 2. The topic under consideration is not the actuality of capitalist production of today but a presumed tendency of capitalist production today, which can be described abstractly as Production by Robots — firms of workers making robots making robots making consumer products competing against rival firms of workers making robots making robots making competitive consumer goods.In this form of production, not only is Hardy's "final" labour displaced by "valueless" robotic labour but his "preliminary" labour is also displaced by "valueless" robotic labour. This annuls Hardy's argument, at least if the displacement of labour by robotics occurs equally throughout both "preliminary" and "final" labour, which includes an equal reduction in the human component embodied in the production of the robots as well.Unlike Hardy's realistic version of capitalism, this form of production may not actually exist, or at least it may not currently predominate. But it is worth considering because it is a theoretically possible implementation of capitalist production, and its preconditions are being generated naturally by current production. [Of course, Marx and Hardy were aware of this tendency toward full-stage, rather than just "final" stage, automation occurring in their own lifetimes.]We defer consideration of the countervailing forces that may offset these preconditions until Observation 3 below.Recursion simply means using A to produce A to produce A … In our case, robots to produce robots to produce … Recursion going on and on forever is only possible for Mother Nature. In actual human production, we only need to swamp costly human labour by cheap robotic labour to approximate humanless production — to approximate a valueless condition of capitalist production.Valueless capitalist production would, of course, be catastrophic for capitalism and for the capitalist class, as a capitalist production process only generates value through human labour and not through robotic labour, which [if truly humanless] is valueless in Marxian terms.Valueless capitalist production would, of course, also have dire consequences for the working class, since labour power would become socially valueless. The working class in such a capitalism would be rendered socially jobless and so socially useless, like the free proletariat of Ancient Rome [that was the same drone class that had the leisure to spread its comforting delusion Christianity].If social labour at all stages of capitalist production ever becomes predominantly robotic, then consumer products become predominantly valueless. In a competitive capitalist economy — one which through competition enforces strict value production — the prices of consumer goods would plummet. Could this happen?Observation 3. What do we make of all this?The consequences of whole-scale robotic production under capitalistic conditions rely on whether:1. it is actually possible to predominantly robot-ize all stages of commodity production,2. it is actually possible to sustain a competitive capitalist market for goods and services provided predominantly by whole-scale robotic production.It is quite possible that the answer to both questions is No! In which case, we need to understand why, e.g. preliminary production may always remain human intensive, unlike final production, presumably through increasing difficulty of obtaining resources without predominantly human involvement.But if the answer is yes, then the problem remains of whether the working class is any longer necessary to capitalism as inactive producer and so unnecessary purchaser of consumption goods — whether the working class in such a capitalism has forfeited its social right to its social existence. It's hard to conceive of a competitive capitalist market surviving whole-scale robotic production. Something like what YoungMasterSmeet raises as a prospect may come to the capitalist market's rescue — but his vision is of a class-monopolistic artificial market, to the dismay of the great defenders of a free and open market.Observation 4. Capitalism is a survivor. It is a supreme example of an adaptive self-organising system, what Marx and Hegel called an organism. We are the agents of its adapting and self-organisation — its perpetuation as well as of its abolition.So far capitalism has met all challenges, even if it is we who, unwittingly, save it. It could conceivably survive this presumed robotic crisis with our help. But should we save it?Capitalism, as a social system, uses us to maintain its daily sustenance and ensure its survival. It has long since ceased to deserve our allegiance. We must consciously put it out of its [and our own] misery.
The Article wrote:“The relative advantages of humans and machines vary from one task to the next. Imagine a chart resembling a topographic cross section, with the tasks that are ”most human” forming a human advantage curve on the higher ground. Here you find chores best done by humans, like gourmet cooking or elite hairdressing. Then there is a ”shore” consisting of tasks that humans and machines are equally able to perform and, beyond them an ”ocean” of tasks best done by machines. When machines get cheaper or smarter or both, the water level rises, as it were, and the shore moves inland.” (“Economics Of The Singularity” (link))Specifically, it's not robots building robots, but robots designing robots that puts the dagger to the heart. (Hanson's article is worth a read too).To put the contrary case to ALB, in the 1960's the Wilson government had to seriously debate whether the economy could afford to raise the school leaving age to 15, in terms of lost workers. The Blair government tried to create an effective school leaving age of 21 (for half of all children). Look in agriculture and just how few are employed in what was once the most labour intensive of industries (and, as I saw someone say recently, look at all those Discovery channel documentaries on "How it is made" to just see how automated production is. Look at "Self-service" in supermarkets, one person works four or five tills now.I went round Cambridge Press's print works last year, they take up half the space they used to, and the handful of staff keeping the presses turning wouldn' take that much replacing (especially if you could create expert systems to do the design/layouts and editting).The summer school version of the talk, I used the analogy with Gold. When party speakers were defending the labour theory of value, some objectors would say a lump of gold found with no effort would have immense value. Our writers/speakers replied, that that value included all the time spent looking and not striking gold. So the digital economy goes, people write programme, practically for love, and some get lucky when their design proves popular.Those who strike gold will need services (servants, assistants, etc.), and in turn they will need services as well, so the growth of the service industry is in line with the rise of automation (and may, even, be an index of it).
And just for ego sake:http://www.youtube.com/watch?v=pjPpcP9hkpoThat's a varient version of the talk (I think a bit more detailed on the question of Singularity)
twc wrote:In actual human production, we only need to swamp costly human labour by cheap robotic labour to approximate humanless production — to approximate a valueless condition of capitalist production.Valueless capitalist production would, of course, be catastrophic for capitalism and for the capitalist class, as a capitalist production process only generates value through human labour and not through robotic labour, which [if truly humanless] is valueless in Marxian terms.(…)If social labour at all stages of capitalist production ever becomes predominantly robotic, then consumer products become predominantly valueless. In a competitive capitalist economy — one which through competition enforces strict value production — the prices of consumer goods would plummet.You are right. Marx did speculate about this in a passage in the Grundrisse where he says, after discussing the deplacement of direct labour by machines in production:
Quote:As soon as labour in the direct form has ceased to be the great well-spring of wealth, labour time ceases and must cease to be its measure, and hence exchange value [must cease to be the measure] of use value. The surplus labour of the mass has ceased to be the condition for the development of general wealth, just as the non-labour of the few, for the development of the general powers of the human head. With that, production based on exchange value breaks down, and the direct, material production process is stripped of the form of penury and antithesis. http://www.marxists.org/archive/marx/works/1857/grundrisse/ch14.htm )Incidentally, this is (I think) the only passage where Marx uses the word breakdown ( "zusammen") in relation to capitalism. What he is saying is that if capitalism continued long enough it would eventually reach a stage where, thanks to mechanization, an individual product would have so little labour (indirect as well as direct) embodied in it that its value, and so its price, would be virtually zero. At which point the change-over to production to give away or take freely would become a necessity for the survival of society.I don't think that Marx thought that this stage would ever actually be reached, but was merely extrapolating trends discernable under capitalism to their logical conclusion (a bit like he did too with his mathematical demonstration that eventually the rate of profit must fall). Capitalism is nowhere near this stage now over 150 years since Marx wrote and, despite robotics, is still (very) far away from it. Hopefully, the workers will have put an end to capitalism long before it reached this theoretical breakdown point. Marx must have thought so too. Otherwise why was he active as a socialist in his day?Basically, then, we are still living in the stage of capitalism described by Hardy not the end-days of capitalism as extrapolated by Marx in the passage above.
Another link of interest http://singularityhub.com/2012/11/12/1-million-robots-to-replace-1-million-human-jobs-at-foxconn-first-robots-have-arrived/ It seems like there is a possibility that this discussion takes us into the realm of the singularity, where such stuff as nano tech and 3D printing brings the cost of production very close to zero in resources and labour input.
Productivity from Top to BottomHardy is correct to remind us not to forget "indirect" production costs, but he is wrong to dismiss the effects of productivity growth within these not-to-be-forgotten "indirect", or "preliminary", production commodities, which — just like "direct" or "final" products — are also produced for a specialist market, even if that market is a restricted market for capitalist-class productive consumption. All actual markets are equally restricted in scope, but are nevertheless free markets.It is notorious that the producers at the bottom of the production chain squeal loudest about being robbed by those above them. That's the hallmark of their good capitalist productivity. Those at the top of the chain compel the producers of their productive "components" to be productive by switching from one to other component manufacturer based primarily on their perceived productivity.Both Direct and IndirectIt seems to me that capitalist productivity is recursive all the way up and down the production supply chain. This is how the social "coercive laws of competition" operate globally throughout capitalist production.As "social laws of nature" every capitalist manufacturer must obey them like a natural "law of nature" so long as he doesn't absolutely monopolize the market — for it is market competition alone that operates as the coercive force that drives productivity everywhere throughout capitalist production.An Example. The semiconductor industry relies on so many independent "final" parts manufacturers [component manufacturers of wafer chips, digital displays, camera lenses, lasers, microphones, speakers, software apps, …] whose social productive function is primarily to be mere "preliminary" products for productive consumption in the production of "final" consumer goods.Each of these component manufacturers is determined to outcompete the productivity of his equally determined rival component manufacturers — equally as determined as the "final" product manufacturer he supplies his "preliminary" product to.So I assert that no commodities are exempt from productivity gains at rates that are appropriate to their level — whether they be "preliminary" or "final" products. In a production chain, one process's "final" product is another process's "preliminary" product.The productivity embedded in the "final" product at the top of the chain is likely to be shackled by the lowest productivity of the weakest link in the supply chain below it [which I now belatedly see is Hardy's precise point]. The productivity embodied in the "final" product at the top of the chain embodies all the productivity gains along the chain, and so its own productivity is more socially "averaged" than the individual productivities of its components.It's just dawned on me, that this social "averaging" of productivity embodied in the "final" product is precisely what the precient Hardy is using to explain why overall social productivity is relatively low. To the extent to which Hardy's case depends on the 'socially averaged' productivity embodied in "final" products, I am compelled to agree with him, but it took me quite a while to realize that I actually did.But my substantive point remains — increasing productivity gains are compelled to occur at every stage of capitalist production, and to become embodied in every saleable product on the market — in every commodity, whether a component or a whole. To argue that — socially wide — the productivity of "final" products is shackled by weak productivity links in the whole social supply chain beggars belief.Surely capitalism doesn't work [or more correctly, fail to work] in that way?Surely, capitalism doesn't tolerate incompetent productivity sabotage on a grand social scale? Surely the socially "averaged" productivity embedded in "final" products is ever increasingly getting higher!At least that's how I see it.And so I feel compelled to revisit Hardy's problem — if ever-increasing productivity doesn't manifest itself in ever-increasing growth, profit and interest rates, what's actually going on?
One question that interests me is the separation of price and value. It is possible for valueless objects to have a price (since anyone can agree to exchange anything for anything). Antiques, art, found objects, valueless but people may swap them to their hearts content. Say I swap a painting for a rock, for example. Now, if we have two valueless objects exchanged in terms of a third valueless object, we have an extended valueless exchange expressed in a price.Essentially, robots could trade virtual money for virtual products, almost endlessly. Markets could continue without value production, replacing system signals for value judgements. Thus, there is the prospect of markets continuing, and life with it, without descent into machine induced barbarism.Charles Stross, in his novel 'Accelerando' has a joke incident in which someone patents a method of central planning involving shell companies; but the idea does intrigue me, could robots (physical and electronic) eventually lead to a situation in which humans don't need money?
- You must be logged in to reply to this topic.