1

‘WORKING PAPER’ APRIL 2017

Automation – friend or foe?

Unite Scotland (Politics, Research and Campaigns Unit)

Introduction

In recent years, the speed and nature of technological developments has led to talk of a new wave of automation, increasingly termed ‘Industry 4.0’.[1] Advances in technology, especially in robotics, sensors, Artificial Intelligence (AI) and cloud computing, are helping to drive the current wave of automation. Debate over the scale and extent of automation in the global economy is contentious, with widely differing figures for the labour market, in particular job reduction and replacement. Therefore, at the outset, it is important to state that there is no academic, industry or union consensus other than to acknowledge there is a process of deepening automation underway but its implications are at present fully unknown. The potential solutions and strategies designed to address the consequences of automation require careful consideration in areas such as education, skills and the social security system. It is in this context, Unite is sympathetic to exploring the virtues of a Universal Basic Incomepolicy as a complement to the social security system to mitigate the effects of automation, and, welcomes the introduction of trials in Scotland of the approach.

This contribution to generating an informed approach to the salient issues begins by reviewing existing research before discussing the certain aspects of productivity and control as well as a series of union demands articulated by the Unite union. The paper then moves to consider the case of Germany before articulating some policy proposals.

Contemporary research

In a recent report by Acemoglu and Restrepo,entitled ‘Robots and Jobs: Evidence from US Labor Markets’,[2] the authors concluded a more pessimistic outlook on the emerging data in regards to job dislocation, specifically in geographically concentrated areas, as a result of automation. The report analyses the effect of the increase in industrial robot usage between 1990 and 2007 on US local labour markets. From 1993 to 2007, the United States added one new industrial robot for every thousand workers — mostly in the Midwest, South and East — and Western Europe added 1.6. The report identified ‘large and robust negative effects of robots on employment and wages across commuting zones’. It is now estimated that one more robot per thousand workers reduces the employment to population ratio by about 0.18%-0.34% points and wages by 0.25%-0.5%. It appears to be the first study to quantify large, direct, negative effects of robots.The researchers noted negligible employment increase in other occupations to offset the job losses in manufacturing. Consequently, there are large numbers of people out of work, with no clear path forward, especially ‘blue-collar men’ without college degrees. Robots are ascribed as being responsible for up to 670,000 lost manufacturing jobs between 1990 and 2007, while industrial robots are expected to quadruple. The paper adds to the evidence that automation, more than other factors like trade and offshoring that President Trump campaigned on, has been the bigger long-term threat to blue-collar jobs.

Robots affected both men’s and women’s jobsbut the effect on male employment was up to twice as big. In an isolated area, each robot per thousand workers decreased employment by 6.2 workers and wages by 0.7%. But nationally, the effects were smaller, as previously noted, because jobs were created in other places. Employment in Detroit was greatly affected. The researchers identified that if automakers can charge less for cars because they employ fewer people, employment might increase elsewhere in the country, like at steel makers or taxi operators. However, in Detroit, people will probably spend less at stores. So, including these factors, each robot per thousand workers decreased employment by three workers and wages by 0.25%.

In ‘The Future of Employment: How susceptible are jobs to computerisation?’, Osborne and Frey stated that automation could lead to 35% of UK jobs being lost within the next two decades.[3] In a speech to the TUC in 2015, Andrew Haldane, Chief Economist, Bank of Englandsaid that up to 15 million jobs in the UK could be at risk of automation over the same time period.[4] In a later report published by the OECD (2016) titled ‘The Risk of Automation for Jobs in OECD Countries’, it found that instead of 47% of US jobs being replaced by automation, as claimed by Frey and Osborne (2013), the figure is only likely to be 9 % at ‘high risk’, while in the UK the threat to jobs is 10% (see table below). The difference between the aforementioned papers is driven by a task-based approach rather than an occupation-based approach. Frey and Osborne assumed that whole occupations rather than single job-tasks are automated by technology. The OECD reports contrasts with this position stating that:‘this might lead to an overestimation of job automatibility, as occupations labelled as high-risk occupations often still contain a substantial share of tasks that are hard to automate’.

Automatibility by OECD countries

In a March 2017 report by PriceWaterHouseCooper,[5] it was projected the following could arise as a result of automation:

  • Up to 30% of UK jobs could potentially be at high risk of automation by the early 2030s, lower than the US (38%) or Germany (35%), but higher than Japan (21%).
  • The risks appear highest in sectors such as transportation and storage (56%), manufacturing (46%) and wholesale and retail (44%), but lower in sectors like health and social work (17%).
  • For individual workers, the key differentiating factor is education. For those with just GCSE-level education or lower, the estimated potential risk of automation is as high as 46% in the UK, but this falls to only around 12% for those with undergraduate degrees or higher.

However, in practice, not all of these jobs may actually be automated for a variety of economic, legal and regulatory reasons. Yet, the research does lend credence to the argument that working-class, lower educated individual workers and in specific industries often geographically concentrated such as automakers, are more susceptible to automation.

Productivity: a crisis of investment

A further factor to consider in the UK in relation to the speed of automation arises from investment and associated levels of productivity. The Fraser of Allander highlighted that productivity growth, particularly in Scotland, where there is limited opportunity to boost performance through population growth or in the labour market, becomes even more pressing to economic growth. The Institute in relation to productivity noted the following: ‘If we can produce more output (or better quality output) whilst still working the same hours then we will be better off. It is therefore an important indicator of economic performance and a key driver of long-term living standards’.[6]

Peat, in a recent paper highlighted the lack of productivity growth in the UK, argued:

Until around the time of the global recession productivity across the UK grew steadily year-on-year by some 1½ to 2%, a rate comparable with many international competitors. Since perhaps the mid-2000s productivity growth has slowed both in absolute terms – implying slower trend growth of GDP per head – and relative to those key competitors – implying reducing competitiveness for companies.[7]

In fact, UK productivity is around 18% lower than the G7 average in 2015 (and significantly below the US, France and Germany). The figure was the same as in 2014 and is the widest productivity gap since at least 1991. In this context, recent comments made by the Bank of England deputy governor, Jon Cunliffe,on why productivity has been weak as a direct result of a lack of business investment,were particularly illuminating, he said: ‘The weakness in productivity growth over the last decade has been accompanied by a weakness in business investment. In the 40 years to 2007, business investment growth averaged 3 per cent a year. In the eight years since the crisis it has averaged 1.5 per cent annually’. Cunliffe further noted a recent survey carried out by the Bank of England, which found three-quarters of companies putting investment behind mergers and acquisitions or paying dividends as a priority. The survey also found that companies required a 12 per cent ‘hurdle rate’ on investments as any investment that returned less than this figure was not made.[8]

Therefore, central to the arguments surrounding boosting productivity levels relates to improved business investment but a critical complement of this investing is in people in terms of skills levels and higher wages. As a result, while the process of automation can be decelerated due to a lack of investment, if this situation is to be reversed in terms of investment – which it should - the process of automation will require careful planning because as a recent report by the Scotland Institute highlights:‘If automation goes ahead unmanaged then both then both wages and employment will fall resulting in increased numbers of people trapped into low wages (if they can find any work at all’.[9] Accordingly, this will require a deeper analysis in the future of what we mean by ‘secure’ and ‘rewarding’ work inclusive of new employment protections.

Control of production

A factor often overlooked in the automation debate is the issue of ownership and control. In the age of automation, the production of communication and information technologies is experiencing contradictory dynamics: increasing centralisation, rent extraction and ownership in conjunction with horizontal diffusion in terms of access, zero-costs and self-ownership.As part of this emergent economy some companies have managed to create a monopoly situation where they control access and transmission (Google, Facebook) of information and use this control to extract rents from the productive process. These shifts are creating concentrations of power, control and wealth on an unprecedented scale as the market leaders create barriers and charge rent for access. In relation to the aspect of service delivery, there are substantial implications for workers and workplace regimes in terms of distribution, technology, and storage as control over ‘gateways’ ensures the extraction of rents from those involved in the process of production and its consumption (companies such as Amazon, AirBnB and Uber). As the Scotland Institute (2016) noted, ‘In effect, income that would have been spread (however imperfectly) across producers, employers and service providers is being concentrated by a few organisations that can extract rent from an otherwise near cost-free distribution system’. Furthermore, we are witnessing these aforementioned platforms/companies pursuing aggressive approaches towards their workers in tandem with tax avoidance measures. As a result of the process of automation inclusive of artificial intelligence and cloud computing, regulatory authorities must address growing concerns in relation to the concentration of power and rent extraction. A part of the equation must be to address the variegated forms of ‘employment’ status of workers and associated myriad of ‘rights’, which is often exacerbated by new forms of workplace surveillance and monitoring.[10]

The fourth industrial revolution

It took fifty years for the world to install the first million industrial robots – introduced in 1961. The second million will take only eight years.[11] Due to the rapid rate of technological change there is a very real threat of job losses to workers in all areas of the economy and in many different unrelated roles, even if the full extent is not known. Unite has identified more than 650,000 Unite members working in high risk sectors at risk of losing their jobs through automation. By using Frey and Osborne’s methodology, 231,768 of these Unite members could lose their jobs by 2035.[12]

In 2013, ‘Robotics and Autonomous Systems’ (RAS) were identified by the Coalition Government as one of the ‘Eight Great Technologies’. However, in October 2016, the Science and Technology Committee inquiry into Robotics and Artificial Intelligence unsurprisingly ‘found that there was no Government strategy for developing the skills, and securing the critical investment, that is needed to create future growth in robotics and AI’.[13]Continued government inertia on this very important issue is squandering a unique opportunity to implement a comprehensive industrial strategy that plans for the ‘21st century workplace’. Unite insists that such an industrial strategy should ensure that technology is only implemented when it is beneficial for all, with work as the central pillar of society. While it is agreed that automation will change the nature of existing work, Unite does not accept that it must result in inferior work or unemployment.

In this context, Unite is calling for:

  • The extension of collective bargaining in the workplace to include the introduction of new technology;
  • Trade union members at board level to be involved in all decisions relating to the introduction of new technology in the workplace;
  • Regulation for how employee information can be gathered, stored and disclosed;
  • A ‘Future of Automation’ Commission involving trade unions, employers, research councils and academics, tasked with finding potential solutions for the future;
  • A commitment to a programme of re-skilling and up-skilling existing workforces;
  • The creation of new training and apprenticeship schemes reflecting changing job roles.

Unite is also aware of the ongoing debate on the question of a tax on robots. This is a question which must be thoroughly debated. Unite will explore the idea of a specific tax where employers use robotics to replace workers, in order to fund research into new technologies, new ways of working and talent retention programmes. Unite is also keen to explore the potential for a shorter working week without a reduction in pay.Manufacturers in the UK are already starting to operate so-called ‘smart factories’, where networked machines and computers work together to enable organisations to digitally plan and project the entire lifecycle of products and production facilities. Experts in Germany, where investment in automation is high expect widespread adoption in factories around the world in the next decade. Unite notes the increased use of robots and ‘cobots’ in sectors such as automotive and aerospace.[14] In Germany cobots are already used in the Ford factory in Cologne and the Volkswagen factory in Wolfsburg for handling ergonomicallyunfavourableproduction tasks, such as overhead assembly.[15] In the United States, automotive employers spent over $718 million[16] on cobots in 2015 and 2016. Any industrial strategy must acknowledge the potential market for building and maintaining both robots and cobots.[17]It is Unite’s view that an industrial strategy must have the ambition of making the UK a global leader in the ‘Industry 4.0’ revolution. This must include:

  • An examination of the potential for current manufacturing capabilities to diversify through automation;
  • Sector-by-sector study into viability of future investment in R&D;
  • Government investment in R&D to make the UK a leader in robotic production and maintenance.

However, the impact of automation is not limited to manufacturing. Unite members working in the transport sector will be seriously affected by the development of automated/driverless vehicles, particularly in logistics and bus and taxi drivers. Daimler already has autonomous technology in demonstration trucks,[18] while Volvo says that they could have the technology within two years.[19]The Financial Times calculates that a driver’s wage accounts for roughly a third of the total cost of owning and running a lorry.This, plus a growing shortage of qualified drivers in Europe, could provide the incentive for the industry to automate.[20]The transport sector in Scotland employs some 98,000 people in Scotland, who earn £2.7bn (an average of £31,702 per employee).

We must ensure that automation does not merely exacerbate the existing ‘race to the bottom’ of wages, terms and conditions. While automation will change the nature of existing work, Unite does not accept that it must result in unemployment. Unite is calling for employers to create new training and apprenticeship schemes which reflect changing job roles, and also to commit to a programme of re-skilling and up-skilling existing workforces.Unite is calling for Government, at the Scottish and UK levels, and employers to create new training and apprenticeship schemes which reflect changing job roles, to recognise changing skill needs and to commit to a programme of re-skilling and up-skilling existing workforces.

The German example

In this context, Scotland and the UK can learn from the emergent strategies in Germany. IG Metall organises workers across German manufacturing and in response the union sought to position itself at the centre of the industrial changes underway. This included lobbying and working with government to shape regulation for industrial health and safety. The union also partnered with universities in Bochum and Darmstadt to establish‚ Qualification Labs, to propose the new types of jobs and apprenticehips needed for digital manufacturing. The union has also signed Germany’s first ‘Industry 4.0’ company level agreement with engineering firm Bosch-Rexroth at Homburg, Saarbrücken. The agreement gives the Works Council, comprised of elected union shop stewards, oversight into the phrased introduction of new production methods. The agreement also includes safeguards for the collection, storage and use of data.While the agreement between the Works Council and Bosch-Rexroth sets a precedent, IG Metall are also aware that automation will impact workers who are no longer in the traditional workplace. Much like Task rabbit, Uber and Deliveroo in the UK, Germany has seen the rise of the so-called ‘Platform Economy,’ where piecemeal work is outsourced online. In Germany, the website ‘Clickworker’ claims to have a network of over800,000 freelancers offering to do work ranging from data processing to publicity.[21] For manufacturing, this means white collar roles which were traditionally in-house can now be outsourced.In response IG Metall has launched its own online platform; Clickworkers can be made aware of their rights and get support from the trade union. This includes rating and comparing both pay and terms and conditions to prevent the race to the bottom. For example, as some online platforms proactively try and forbid freelancers workers from communicating with each other, the union’s site allows Clickworkers to collaborate and share details on pay.[22]

The fallout from automation: policy proposals

As part of the wider emergent strategies to dealing with the consequences of automation,policy prescriptions such as the Universal Basic Income are gaining political traction. In Scotland, Glasgow and Fife Councils, for example, are currently undertaking exploratory studies to introduce trials. The Finish Government has also drawn up plans to pay every adult citizen a basic income of €800 each month.