De-industrialisation and the prospects for socialism

Is the world really de-industrialising?
Is the world really de-industrialising?

by Michael Roberts

Last week I spoke on a panel that debated De-industrialisation and socialism.  The panel was organised by Spring, a Manchester-based group in England that has become a forum for the discussion of developments in capitalism and their implications for the prospects for socialism (

The main theme for this panel discussion was the evident fact that the industrial sector (manufacturing, mining, energy etc) has declined sharply as share of the output and employment in the mature capitalist economies during the 20th century.  The question for debate  was: does this mean that the working class has also declined and is no longer the main force of change in capitalism; and also that a socialist or post-capitalist society will be a world without industry or employment of industrial workers?

World not de-industrialising

The first point I made in the discussion was that the world is not de-industrialising.  Globally, there were 2.2bn people at work and producing value back in 1991.  Now there are 3.2bn.  The global workforce has risen by 1bn in the last 20 years.  But there has been no de-industrialisation globally.  De-industrialisation is a phenomenon of the mature capitalist economies.  It is not one of the ‘emerging’ less developed capitalist economies.

Using the figures provided by the International Labour Organisation (ILO) we can see what is happening globally, with the caveat that there is a serious underestimate of industrial workers in these figures and such transport, communication and many hi-tech workers are put in the services sector.

Globally, the industrial workforce has risen by 46% since 1991 from 490m to 715m in 2012 and will reach well over 800m before the end of the decade.  Indeed, the industrial workforce has grown by 1.8% a year since 1991 and since 2004 by 2.7% a year (up to 2012), which is now a faster rate of growth than the services sector (2.6% a year)!  Globally, the share of industrial workers in the total workforce has risen slightly from 22% to 23%.  It is in the so-called mature developed capitalist economies where there has been de-industrialisation.  The industrial workforce there has fallen from 130m in 1991 by 18% to 107m in 2012.

Global workforceThe big fall has not been in industrial workers globally but in agricultural workers.  The process of capitalism sucking up peasants and agro labourers from the rural areas and turning them into industrial workers in the cities is not over.  The share of agricultural labour force in the total global workforce has fallen from 44% to 32%.  So should we not really talk about de-ruralisation, as Marx did in the mid-1800s?  That is the great global phenomenon of the last 150 years.

Of course, most workers globally work in the services sector.  This sector is badly defined, as I say, and is really anybody not clearly an industrial or agricultural worker.  This sector was smaller than agriculture in 1991 (34% to 44%) but now it is biggest at 45% compared to 32% for agriculture.


As I was speaking in Manchester, the centre of the industrial revolution in Britain in the early 19th century, I was reminded of the work of Friedrich Engels, Marx’s partner in crime, who was managing his uncle’s German firm in the city at the time.  As a young man (24 years), Engels wrote The condition of the working-class in England (published in 1845) (
and described the horrendous conditions of squalor, disease, sweat shop conditions, injury and poverty that rural men, women and children were subjected to as they came to work in the fast industrialising and urbanising cities of northern England.  It’s the same story now in the likes of India, China, south-east Asia and Latin America.

Engels concentrated on the conditions for labour, but in a preface to a new edition of his book in 1892, he commented that Britain was fast being replaced as the major industrial capitalist power by France, Germany and the US.  “Their manufactures are young as compared with those of England, but increasing at a far more rapid rate than the latter.  They have reached the same phase of development as English manufacture in 1844”.  And so it is now for the so-called emerging economies of Asia, Latin America and Africa compared to the mature capitalist economies of Europe, Japan and North America.

But it’s true that the share of industrial workers in the mature economies has fallen from 31% in 1991 to 22% now.  Indeed, according to McKinsey, manufacturing employment fell 24% in the advanced economies between 1995 and 2005.

US manuf emp

A world of leisure?

So does this mean that the future of capitalism without an industrial proletariat capable of being an agency for change and, for that matter, ‘post-capitalism’, will be a society without industry, where people can expect to reduce their hours of work for a living and have increased periods of ‘leisure’?

This was the theme that my fellow panellist Nick Srnicek posed.  Nick is a Fellow in Globalisation and Geopolitics at UCL. He is the author with Alex Williams of Inventing the Future (Verso, 2015) and the editor with Levi Bryant and Graham Harman of The Speculative Turn (, 2010) – see
Nick explained that, while new economies were being industrialised, their peak of industrialisation came earlier than for economies like Britain in the 19th century.  Indeed, no economy had achieved more than a 45% share for industrial employment.  So the future is not industry and an industrial working-class.  And it was no good advocating a return to manufacturing and industry as the way forward for a better society.

I am sure that Nick is right in these points. Where I differed was that he was not clear if a post-capitalist, non-industrial society would be achieved gradually as capitalism expanded globally and technology replaced heavy industrial work and people worked less hours and could use their time for themselves.  The idea of a steady move to a post-industrial, leisure society was the concept of Keynes back in the 1930s, arguing for capitalism as the way forward to his students at the height of the Great Depression in the 1930s, when many of his students had started to look to Marxism as the explanation for crises and the alternative of socialism (see my post,

Keynes reckoned the capitalist world would achieve huge per capita GDP growth and enter a ‘post-capitalist’ leisure economy without poverty.  Well, this blog has regularly revealed data that shows poverty remains a terrible spectre over the globe, an inherent feature of capitalism, and that far from moving to a leisure society, working hours have hardly fallen much in the mature economies and remain very high in industrial sectors of the emerging economies.  We are all still ‘toiling’ for a living (apart from the 1%), in increasingly precarious jobs.

Need to change mode of production

I don’t think we can get a ‘post-capitalist’ leisure society through gradual change.  It will require a revolutionary upsurge to change the mode of production and social relations globally, even if the potential productivity of labour through new technology and robots etc  is already there globally to deliver such a transition to freedom from toil.  Capitalism remains in the way as a fetter on production, with capitalists as a class force opposed to freedom.

The reason that the mature capitalist economies have lost their industrial base is that it was no longer profitable for capital to invest in British industry in the late 19th century or OECD industry in the late 20th century. So capital counteracted this falling profitability by ‘globalising’ and searching for more labour to exploit.

And profitability fell because capitalist accumulation is labour-shedding.  Capitalists compete against each other to get more profit.  Those capitalists with better technology can steal a march on others by boosting labour productivity and reducing labour costs by cutting the workforce.  So the drive is always for reducing the amount of labour power to boost profits.  The central contradiction here, as explained by Marx’s law of profitability, is that the reduction in labour power relative to mechanisation leads to an eventual fall in profitability.  This reduces the industrial workforce in the mature economies and leads to expansion of industry globally.

Capitalism is a mode of production for mechanisation, but mechanisation will also lead to its demise as it is a mode of production for profit not social need and more mechanisation eventually means less profitability.  That shows that as we move towards a robot economy: profit for capital and meeting social needs will become more incompatible.  And the leisure society just an impossible dream.

Employment growth is falling in the advanced capitalist eocnomies. Employment growth is way less than 1% a year in the 21st century.

AD empComputer engineer and Silicon Valley software entrepreneur, Martin Ford puts it this way: “over time, as technology advances, industries become more capital intensive and less labour intensive.  And technology can create new industries and these are nearly always capital intensive”.  The struggle between capital and labour is thus intensified.

Class struggle

It does depend on the class struggle between labour and capital over the appropriation of the value created by the productivity of labour.  And clearly labour has been losing that battle, particularly in recent decades, under the pressure of anti-trade union laws, ending of employment protection and tenure, the reduction of benefits, a growing reserve army of unemployed and underemployed and through the globalisation of manufacturing.

According to the ILO report, in 16 developed economies, labour took a 75% share of national income in the mid-1970s, but this dropped to 65% in the years just before the economic crisis. It rose in 2008 and 2009 – but only because national income itself shrank in those years – before resuming its downward course. Even in China, where wages have tripled over the past decade, workers’ share of the national income has gone down. Indeed, this is exactly what Marx meant by the ‘immiseration of the working class’.

What about the robots?

Will it be different with robots? Marxist economics would say no: for two key reasons.  First, Marxist economic theory starts from the undeniable fact that only when human beings do any work or perform labour is anything or service produced, apart from that provided by natural resources (and even then that has to be found and used).  So, crucially, only labour can create value under capitalism.  And value is specific to capitalism.  Sure, living labour can create things and do services (what Marx called use values).  But value is the substance of the capitalist mode of producing things.  Capital (the owners) controls the means of production created by labour and will only put them to use in order to appropriate value created by labour.  Capital does not create value itself.

Now if the whole world of technology, consumer products and services could reproduce itself without living labour going to work and could do so through robots, then things and services would be produced, but the creation of value (in particular, profit or surplus value) would not.  As Martin Ford puts it: “the more machines begin to run themselves, the value that the average worker adds begins to decline.”  So accumulation under capitalism would cease well before that robots took over fully, because profitability would disappear under the weight of ‘capital-bias’. This contradiction cannot be resolved under capitalism.

We would never get to a robotic society; we would never get to a workless leisure society – not under capitalism.  Crises and social explosions would intervene well before that.

The above article first appeared on Michael’s excellent The Next Recession blog, here.

Further reading:
Whatever happened to the leisure society?
Pensions and the retirement age: the problem is capitalism, not an aging population


  1. Failing to see capitalism in its totality leads people to say a lot of crap about changes in the way the system functions and this doesn’t just include ignoring the off-shoring of production, but also the circulation of capital in developed countries. There’s a strong fetishisation of changes in consumer society, but Marx clearly lays out in volume 2 of Capital that circulation is predominately business to business, rather than business to consumer. I experienced this myself while working for a Barter firm (BarterCard), the clients that I managed were predominately biz-biz enterprises and they tended to fair a lot better than the majority of small consumer businesses who struggled to keep their heads above water.

  2. One of the critically important insights of Marx was how people form their ideas based on the surface appearances of the system and their (inevitably) partial experiences of its operations. These are real, so they create a solid belief in people’s minds. So, as you say, fetishism – in the sense that Marx used the term, not fetishism in the desire sense! – is incredibly powerful.

    One of the things that we have to do is work out how to break through this barrier, because it’s not built on nothing or on simple bourgeois propaganda, as a lot of the left seem to believe – as if the working class had empty heads that simply sucked up the bourgeois ideas beamed at them by the capitalist media.

    It’s interesting when people who know some solid Marx work in an area like finance, because you can see right in front of you how that part of the system works but also know that, while real, this is also surface stuff and that beneath it is a whole other level of the operation of the laws of motion of capital.

    I try to deal with how capitalist ideology works here – – but that article just provides bare bones framework, elaborating the theory of ideology that is in Marx (and was further developed by Lukacs, an absolutely key thinker in my view). I have been thinking for some time that I need to expand that article by giving concrete examples from society today, especially around workers’ direct experiences and stuff like bourgeois arguments in relation to the minimum wage, low pay etc.


    • Commodity fetishism is probably one of the most misused and abused terms of Marx. I’ve experienced this abuse most prominently in media studies papers, where the idea of the relations between Things taking the place of the relations between people is described in a deterministic manner rather than an ideological process. This then leads to the so called counter-argument against Marx, that just as in pre-capitalist society where objects were used to facilitate in strengthening social bonds, so to do commodities (Face palm). Don’t get me started on use-vale referring only to actual function and exchange-value referring to some kind of symbolic value. I really do feel sorry for people who are exposed to such vulgar interpretations of Marx in universities and gain a false sense of understanding, which they then use to argue that Marx was wrong.

  3. It’s also interesting that good academic practice suggests reading primary sources, yet lectures which include stuff about Marx *usually* avoid Marx and cite the professional marxologists instead. Some marxologists are very good, but many of them are simply awful. So crappy secondary sources are used and recycled by lazy academics and crappy marxologists and then the next round of people repeat those sources and on and on until their ridiculous misinterpretations of Marx become some kind of standard *knowledge* of Marx.


Comments are closed.