“Hon Hai has a workforce of over one million worldwide and as human beings are also animals, to manage one million animals gives me a headache”
– Hon Hai chairman, Terry Gou.
[dropcap size=big]I[/dropcap]t’s not a subject you’ll hear many politicians or even economists waxing lyrical about. Structural unemployment in a global sense, put simply, is the declining need for human labour and skill due to increasing technical efficiencies. These efficiencies are almost exclusively technological in nature and represent Moores law in action. Moores law was first penned by Gordon Moore, the co-founder of the company Intel in 1965. He witnessed the number of circuits inscribable per square inch on integrated circuits doubling every year and predicted this trend would continue into the foreseeable future. This trend slowed slightly after a couple of decades but has maintained a doubling every 18 months for well over twenty five years and is expected to remain at this rate for at least two more decades.
This phenomena parallels the more recent development of convergence science, where many different scientific sectors like molecular biology, immunology, molecular engineering and nanotechnology are melding and creating new scientific offshoots through their interdisciplinary copulation. An example might be where engineering has moved into areas as diverse as human biology and quantum computing. The resulting fruit of these new unions in science is vast leaps in automation and AI or artificial intelligence. Leaps which will rewrite the fabric of human society over the coming decades. The processing and manufacturing power of the global economy is set to shift almost an order of magnitude as these technologies proliferate and innovate in every direction rapidly and radically increasing the scope and intensity of human wealth and knowledge creation.
But the flip side of this scientific convergence leading the current automation revolution is that industrial efficiency and human needs from the same changing economy are not always acting in concert. Job losses are the inevitable result of this automative efficiency and this has the capacity to create the conundrum of efficiency shrinking the consumer market as unemployment restricts more and more of the population’s participation in those same global markets.
Recent estimates suggest that tens of millions of Chinese workers will be replaced by millions of robots over the next two decades. Both their contribution to public wealth and their collective consumption’s boost to economic growth and GDP will decline starkly as cheap, inexhaustible, non striking robots replace 70 % or more of manufacturing jobs in China. And signs indicate this process has already well and truly begun.
Foxconn, Apple’s iPhone manufacturer, continues to build new plants and hire thousands of additional workers to make smart phones – but it plans to install more than a million robots within only a few years to supplement its work force in China. If this transition is successful Foxcom, claims this process of automation will continue as it expands operations. Foxconn has not disclosed how many workers will be displaced or when. But its chairman, Terry Gou, has publicly endorsed a growing use of robots as being an option to what he described as ‘the headache of dealing’ with one million workers.
Workers which he also described as animals which perhaps hints that for at least some of the executive class driving this innovation, human workers are being seen increasingly as liabilities. Here’s the quote by the Hon Hai (Foxconn) CEO,
“Hon Hai has a workforce of over one million worldwide and as human beings are also animals, to manage one million animals gives me a headache,” said Hon Hai chairman Terry Gou at a recent year-end party, adding that he wants to learn from Chin Shih-chien, director of Taipei Zoo, regarding how animals should be managed.”
This ‘disruptive’ technology of automation which CEO’s like Terry Gou are aggressively pursuing are part of the next big swing in labour practice since the new international division of labour began when large corporations began seeking cheaper labour markets in the 1970’s. A shift that saw millions lose their jobs in developed nations manufacturing sectors as whole industries moved their operations to zones of far cheaper labour. And as it drops costs and raises profits for the business sector while diluting demand for labour, it is almost certain to become even more disruptive to the existing societal framework.
This is gonna sting a little.
Erik Brynjolfsson, a professor at the MIT Sloan School of Management and co author Andrew McAfee wrote the best-seller Race Against the Machine. They illustrate the point that education investment is the key to staying ahead of the technological changes that automation is bringing about. They talk about Skill-biased technical change and give historical examples of how agriculture and industry through the 17th to 19th centuries adapted to machines age developments by increasing the educational and learning capacity of its citizens via direct investment into education.
“Skill-biased technical change has also been important in the past. For most of the 19th century, about 25% of all agriculture labour threshed grain. That job was automated in the 1860s. The 20th century was marked by an accelerating mechanisation not only of agriculture but also of factory work. Echoing the first Nobel Prize winner in economics, Jan Tinbergen, Harvard economists Claudia Goldin and Larry Katz described the resulting SBTC as a “race between education and technology.” Ever-greater investments in education, dramatically increasing the average educational level of the American workforce, helped prevent inequality from soaring as technology automated more and more unskilled work. While education is certainly not synonymous with skill, it is one of the most easily measurable correlates of skill, so this pattern suggests that demand for up-skilling has increased faster than its supply.”
This investment in training and education systems made the displaced workers able to shift into other emerging industries as their own labour was replaced. Literate, numerically trained workers were more easily and quickly trained to slide into newer more skill intensive positions. But the authors also claim the current up-skilling of the workforce by many OECD nations including the US is not keeping up with demand. Brynjolfsson and McAfee argue that even were up-skilling keeping pace with demand, there will still be winners and losers and ‘some pain’ is unavoidable for much of the population making the transition to a new economic balance. This almost Darwinian approach has a ‘winner take all’ view that sits uncomfortably in contrast with the developed worlds cultural mores of social justice, equity and fairness. It could certainly be argued that this is simply a capitulation by the authors to historical precedent and present day economic rationalism. Something that will not do well in this modern era’s information rich and more dissident ridden milieu.
But even so the point that education and training for highly skilled jobs is not keeping up with the rapidly increasing demand is both compelling and worrying as the authors argue that these will be essential for the least painful transition into the automated world.
It’s all gonna be ok, maybe.
Cory Doctorow, the best selling novelist and columnist for Science Fiction Age and tech writer for the O’Reilly network is more of a technological optimist and suggests that as robots drive down production costs, the necessity to manufacture in far away nations with cheaper labour will disappear. Local production will become cheaper as transportation costs will not drop anywhere near as much as production costs.This will in turn invigorate support industries of robotic manufacturing in delivery, software and maintenance areas to name just a few. He also argues that much of the work replaced by robots is not ‘good work’ but rather dangerous boring and monotonous work that is not possible or particularly safe or done well by humans. That the bulk of jobs performed by automation so far is actually work that humans simply cannot do – or at least do at the current requirement of our hyper kinetic global economy.
“While the displacement of formerly human jobs gets all the headlines, the greatest benefits bestowed by robots and automation come from their occupation of jobs we are unable to do. We don’t have the attention span to inspect every square millimetre of every CAT scan looking for cancer cells. We don’t have the millisecond reflexes needed to inflate molten glass into the shape of a bottle. We don’t have an infallible memory to keep track of every pitch in Major League Baseball and calculate the probability of the next pitch in real time. We aren’t giving “good jobs” to robots. Most of the time we are giving them jobs we could never do. Without them, these jobs would remain undone.”
Furthermore, he argues that the jobs which we didn’t know we needed done are only revealed by the robotics revolution as these new technologies employment creates whole new possibilities that never existed before their antecedents came on the scene. For example, the Google search engine and SEO (search engine optimisation) didn’t exist before the internet, which in turn couldn’t exist before the home computer, which itself only came about because of the development of the micro transistor by William Shockley in 1948. The modern digital transistor was only commercially possible because of new manufacturing methods that made it developed by 1963. This ‘causal chain’ sells the idea that we cannot possibly see all the potential industries that Doctorow believes will emerge in the coming decades to ameliorate the tide of unemployment created by robotics and AI. He also thinks that in order to protect their own markets, some of the profits of automation will have to be redistributed perhaps via taxation by the private sector back into education, training and welfare to offset the impacts of automation.
“The real revolution erupts when everyone has personal work-bots, the descendants of Baxter, at their beck and call. Imagine you run a small organic farm. Your fleet of worker bots do all the weeding, pest control, and harvesting of produce, as directed by an overseer bot, embodied by a mesh of probes in the soil. One day your task might be to research which variety of heirloom tomato to plant; the next day it might be to update your custom labels. The bots perform everything else that can be measured.”
“Right now it seems unthinkable: We can’t imagine a bot that can assemble a stack of ingredients into a gift or manufacture spare parts for our lawn mower or fabricate materials for our new kitchen. We can’t imagine our nephews and nieces running a dozen work-bots in their garage, churning out inverters for their friend’s electric-vehicle startup. We can’t imagine our children becoming appliance designers, making custom batches of liquid-nitrogen dessert machines to sell to the millionaires in China. But that’s what personal robot automation will enable.”
Even in the military arena, hyper fast drone fighter planes and drone robots for all manner of theatre applications are in design in prototype workshops, diminishing future human roles in warfare. Which is something which has its own serious implications for further dehumanising human conflict via remote operation robots and eventually even giving the power to completely AI operated machines to kill human beings. Automation promises to create further unique social quandaries other than mass unemployment, as the types of roles automation takes over pose new unprecedented ethical concerns for regulators.
So do we take the above views at face value? We can assume that automation is not going to occur without some social pain but its seems plausible that automation technologies economic disruption could be greatly mitigated with careful management, suitable investment and forward thinking by both the private sector and governments. Nor are all jobs going to be as easy to automate as others so the process will be perhaps slower, at least in some areas, than we may initially anticipate.
The question is, do we see the kind of long term planning necessary to mitigate the negative effects of massive global automation happening and are claims that a net job loss is not going to result from global mass automation in coming decades likely? Or is it just wishful thinking? Will the new industries created by technological change ‘always’ replace the jobs lost or will the growing efficiencies in automation phase out more and more economic demand for our labour?
The problem here is just how difficult it is to actually accurately measure such statistical realities – let alone projections into the future for changes which have not yet fully manifested. An example of this difficulty is found in getting past the legal privacy around accounting. It is often very difficult to establish the relationship between reduced business income and employment lay-offs in a statistically meaningful way. So how much harder again a task will to predict if we can retain jobs for everyone and not suffer net losses in global employment over the next two decades. Especially when we consider the possible range of sweeping changes to both business income and employment requirements over the next twenty five years. Those kinds of projections are both fraught with unknown variables and the spectre of Murphy lurking ever present in the background.
It would seem prudent then from a policy forming position to assume that there ‘may’ be a serious net loss of employment as the worlds population continues to grow and the cache of available jobs decreases with every improvement in automation technology. Nor will this technology be restricted to blue collar jobs as artificial intelligence and more sophisticated software and networks are able to increasingly replace white collar professionals particularly in the financial industries. Simply put, outside of hi-tech and creative intensive industries, it’s likely the automation revolution will ubiquitously replace unprecedented swathes of white and blue collar workers alike.
As a result of this ratio change of demand for labour versus population growth, its plausible that the very availability of work will become rarer in the future. Perhaps even enough to eventually make work itself a valuable and even trade-able commodity. The mind shudders at the prospect of a job ‘stock market’ where workers will have to compete for employment within a trading structure that makes them little more than corporate serfs. Certainly this kind of darker outcome of the mass automation of global employment falls in line with the law of unintended consequences, though the more conspiratorially inclined reader might suggest it was an engineered result.
And how will society deal with far higher levels of unemployed and fund the safety nets to provide for the vast numbers of displaced workers, many of whom may never work again. Certainly the private sector will need to acknowledge in their future practices that they cannot claim all the benefits of automation without a matching parity in their fiscal responsibility for the costs. Not without collapsing the very markets they depend on for future revenues in a kind of ‘robbing peter to pay paul’ that isn’t remotely sustainable in the medium, let alone long term.
Another ‘New Deal’?
President Franklin Delanio Roosevelt, after the Great Depression, created a program called the ‘New Deal’ between 1933 to 1936 to protect the US from such a severe economic calamity again. This included a strict regulatory framework, including the famous Glass – Steagall Act which, until the Reagan ‘reforms’, prevented banks from speculating with their depositors money. The ‘New Deal’ also included the construction of a social welfare safety net and access to medical services which had not existed before. This “Third industrial revolution” as Jeremy Rifkin labels it in his book of the same name, driven by massive uptake of automation technologies, may well require even more comprehensive ‘New Deal’ type reforms in most nations.
Certainly, in contemporary terms, wealth inequality is increasingly occupying public conversation as statistics show OECD nations slipping with alarming speed back to the late 19th century Downton Abbey era of gross wealth disparity and appalling conditions for most workers. Automation will almost certainly compound this current trend of wealth disparity which according to economists like Joseph Stiglitz and Thomas Piketty threatens the stability of most economies over the long term. And this gross inequalities pressure on consumer markets and employment is likely, eventually, to force currently reluctant governments to re-examine taxation measures and corporate responsibility for the second mass re-engineering of the labour market in recent decades.
Such a ‘New Deal’ would need to officially recognise that automation in our neo-liberal capitalist economy has previously been only used when it cuts costs and reduces the labour needed to maximise profit. An automation system that is more expensive to build and maintain than human workers is ‘never’ used to ‘free’ workers from labour and give them more time for better activities. It is used to increase bottom line profits while the now unneeded workers are divested of previous income. So in net terms, the recognition will need to come that automation technologies will likely involve less workers being employed, and those who will be employed around automation will be in speciality jobs requiring higher degrees of training and experience. Training which, unfortunately will be beyond the reach of many of the people who are generally replaced by such systems.
For the retrenched there is the cost of retraining and the earning time lost while doing so. People who have families to feed and mortgages to pay cannot spend time unpaid taking college or technical classes to up-skill. The time and money for retraining is a luxury many are likely not to have so they will have to find new existing work instead of adapting their skills to the changing jobs landscape. The painful caveat being that there will be far less of those remaining old era jobs for them to find. So for a ‘New Deal’ set of reforms retraining, income support and education would need to be ‘far more’ massively funded to both adapt and occupy meaningfully the many who will lose employment.
Such social reforms would also need to recognise that displaced workers are usually older people who are often less educated, but earnest, honest, and hard working. They are also people usually in poorer health who often don’t learn new things as rapidly and they don’t adapt as well to change. So becoming jobless at fifty years of age in coming decades will almost certain become a sentence to poverty and unemployment due to the already rampant ageism in hiring practices. So what possible solutions and sweeping reforms would we have to find to deflect this rather grim looking future outcome.
In Switzerland there is a referendum in July that seeks to provide a basic wage for everyone ‘whether employed or not’, to leave no one in poverty and to combat rising wealth inequality. All adults receive this basic income regardless of their employment status as a replacement to complex tiered and insufficient welfare payments. As a solution this ‘Basic’ income would almost certainly be part of many nations reforms to prevent a collapse in domestic spending and subsequently of the consumer market and the rise of serious social unrest. When added to extensive education programs, freely provided, this could allow the mass populi up-skilling Brynjolfsson and McAfee talk about while a higher level of income support would allow domestic spending and quality of life standards to stay at sustainable levels. Of course this would be incredibly expensive and would require substantially increased tax based redistribution from the upper economic quintile and a sharing of the capital benefits of automation within society in general. But of course, ignoring its multi generational effects could be far more expensive in the long term.
The growing amount of distrust the general population is already developing towards innovation and is something a ‘New Deal’ type reform package would also have to deal with effectively to gain public support. Recent generations, says NESTA the UK institute for innovation, have developed an increasingly qualified feeling about innovation. This trend over the last two decades is statistically pronounced and NESTA’s research paper “Speaking to the Innovation Population” shows only a third of their respondents thought that innovation was intrinsically valuable. This startling level of societal distrust will be a major side effect of future societies reeling from the disruption of mass automation and commercially viable artificial intelligence.
NESTA’s study showed the largest profiled group making up over half the population liked innovation but only if the benefits and costs were spelled out clearly. Despite greatly appreciating technology and its benefits they were also worried about waste, pollution , social distancing and unemployment as side effects of new technology. This growing suspicion in relation to innovation is the result of a perception of often polarised benefit that often favours the capital investors over the labour force. Add in the dire lack of public discussion of how new technology is going to reshape the economy pro and con and it is easy to see where this growing public cynicism is coming from. It could certainly be argued that any globally successful attempts to ‘head off’ the more calamitous effects of automation and AI will involve getting ahead of the public education curve and establishing pre-emptive policy reform to deal with the inevitable negative effects of technological automation in coming decades.
The defensive contention by ‘market solutions’ advocates that ‘economy and society’ can be flexible enough to find interesting work to be done for each person is certainly an optimistic appraisal in the face of massive likely job losses to automation. When and how that seamless adjustment will take place would be a good question to ask. Few economies are providing those opportunities now or moving with any alacrity to deal with the rapidly increasing wealth disparity already existent since the massive off shoring of labour and manufacturing began in the 70’s.
So how would the enormous necessary structural changes to deal with ubiquitous automation in the future be effected as the public tax base shrank and more wealth than ever was concentrated in the hands of a smaller elite group. Certainly there is a clock on this process as this growing wealth disparity alongside other factors like ageing populations and declining birth rates will soon begin to drastically effect governments revenues. Which will in turn impact on their ability to control their own economies and regulate the increasingly wealthy and powerful corporate conglomerates.
And considering the majority of technological innovation and ideas have birthed from public funding and educational organisations before being exploited by the private, sector it’s not simply a case of ‘fair gain’ via the private ownership of automation, but also a ‘fair return’ to the general public for their long term legacy of primary investment through ‘public funds’ and intellectual capital. The sum total of historical Investment via public universities and government sponsored research foundations as the primary contributors, have been what has made the automation revolution fundamentally possible. This of course will require some rapprochement.
The truly frightening thing that emerges from examining the range of possible outcomes is that there is little evidence that public officials, politicians and the executive class of the private sector have even ‘begun’ to adequately and publicly discuss the impact of automation and near future technologies and their impact on the future landscape of employment, economic growth and wealth distribution. Let alone make cohesive multinational long range plans to compensate for much higher unemployment, shrinking consumer markets and how they might distribute the benefits of automation more evenly enough to prevent the global economy from becoming so lopsided that it topples into collapse .
Perhaps this is the effect of relatively short electoral cycles and quarterly economics, but its seems that current holy sacraments of current socio economic thinking will need to be seriously challenged. And challenged they must be before we can effectively deal with the rapidly approaching automation conundrum and prevent its negative effects from shattering our delicate socio-economic balance.
In the long run, much like humanities other major challenge, Climate Change, early reform and preparation is likely to be the far less painful option than a conservative or delayed action. Time will tell.