EPQ draft 1 (4844 words)
Introduction
Automation is set to un-employ people at a scale and rate never seen before, while simultaneously changing societies very nature on an epic scale; to mitigate its impact we must undertake projects and policies that push ourselves, humanity, and society to its limits. The future could take one of two shapes; a utopian wonderland where everyone is happy, or a dystopia where algorithms and machines run the world for maximum efficiency leaving humanity in the slums. However, despite the impending danger, we are seemingly unaware of it. This is because automation creeps in slowly, so we don’t notice it, only once the teamsters, lawyers, and the CEO’s start to lose their jobs will we notice the extent of what we have created. This is nature of automation you don’t notice until it’s your job, your income, your life that is affected.
However, we might hope that our governments have foreseen this danger and are planning how to avoid it. Yet we do not see any real indication from the government (as of time of writing) that they need to do anything to prepare for the unemployed masses. On the contrary in the 2017 autumn budget the government announced that they wanted to see fully driverless cars by 2021 (HM Treasury, 2017). This is even though a CGPS study showed that over 4 million jobs will be lost to driverless cars in the US (Center for Global Policy Solutions., 2017). IN this report I aim to investigate three main key factors that need to be addressed by both policy makers and the public, to prepare ourselves for the future. These are; Too what extent will automation occur, how will this automation effect society and how should we mitigate its impact? By answering these three questions I hope to bring further clarity to a pressing issue that will affect society from top to bottom.
Too what extent will automation happen
The first question that must be answered in this topic is how much automation will occur. There have been many studies that have attempted to estimate this, and their findings have varied considerably. The most recent study is one by PWC (Berriman & Hawksworth, 2017). By analyzing the other studies and improving upon their methods, they conducted a new survey which concluded that.
“Our analysis suggests that 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%).” (Berriman & Hawksworth, 2017) I have converted this data into a graph below along with other recent studies.
As we can see from the graph the results of automation studies vary greatly. The original study by FO classified jobs as automatable or not by looking at whether most of its tasks could be automated, this meant that they could develop an algorithm to predict what jobs could and could not be automated. AGZ on the other hand, claimed a job was only automatable if all its tasks where fully automatable. This however leads to a vastly reduced number of jobs being classed as automatable. If a job has ten parts and five can be automated surely you can still fire half the workers and maintain the same output. This meant that the results of the AGZ study underestimate the likely impact of automation. For this reason, I have chosen to use the most recent study (PwC) to base my study off.
Will new jobs be created?
If we can have expected jobs to be lost surely, we should also expect new ones to be created to take their place? However, despite this logic the data suggests otherwise see the graph below (Durden, 2017).
As we can see quite clearly although rig count and therefore oil production has increased, the number of employers has stayed almost the same. This indicates that we are producing more oil with less and less energy required. What’s more interesting if you look at the percentage of eligible workers employed over time (Gross, 2016).
As you can see from the graph, the percentage of people employed increases until a recession; at this point employment drops as business make cuts and increase automation, employment regains and then falls. Overtime this cycle reduces peak employment. Furthermore, we can see that the greater the recession the more jobs are irretrievably lost. Currently many believe that we could be in one of the biggest bubbles of all time. The crypto bubble. This term refers to the presumed bubble of cryptocurrencies such as bitcoin and Ethereum. The graph below shows the price of bitcoin over the last year alone (Coindesk, 2017).
It is widely assumed and even accepted that the crypto bubble will be the biggest of our lifetimes and that at some point it will almost certainly burst crashing everyone else with it. This conclusion is mainly drawn from its similarities to the dotcom bubble which burst in 2000–2002. In comparison to the crypto bubble the dot com bubble is expected to look almost reasonable. However, many technology enthusiasts point out that this is okay seeming as we do now all use the internet (Katz & Verhage, 2017). Either way if this is a bubble then we should expect to see job losses in the extremes, jobs that we should not expect to return.
Finally, we must consider whether the new sectors will produce new jobs. Technologists often argue that despite jobs being automated its fine because new jobs such as software developers are being created. And yes, this is true. Whilst it would be unthinkable for the jobs ‘Computer Software Engineer’ or ‘Computer Programmer’ to have existed back in 1980 and now there is 1,300,000 of them. Does not deny the fact that one team of software engineers eleven can design, build, and deploy the next ‘killer app’ within two years and walk away with one billion dollars from its sale. This is the story of Instagram (BBC News technology, 2012). This is a classic case; one we can expect to see much more of. It demonstrates that you no longer need tons of workers to make tons of money. So yes while a few new jobs will be created, we should not rely on these new jobs to support people.
In conclusion we can expect to see lots of job losses within the next 30 years. This is due to the combined effect of automation enabling more jobs to be automated and an impending recession that will drive business to make cuts and improve their business efficiency, with an increasingly small number of people required to make a business successful. We can expect to see unemployment rise to 30% by 2030 and possibly even as high as 50% by 2050.
Why do we work?
Throughout all of time humanity has been in a constant struggle to survive. From stone age man hunting the mighty mammoth, to office workers hunting the mighty pay rise. Humans have always had to strive to survive. Never have we been given the opportunity to simply have it provided to us, although yes, we do get our sustenance easier now than ever before we still must work to get it. But what if we didn’t?
There are three main points of view on the meaning of work. Some people believe that we work because it gives us meaning, that without it we would be aimless with no purpose. The other group say that we work simply for the money and that if that wasn’t an issue we could quite happily live our lives doing what we really wanted to do.
Some people argue that we work to give meaning to our lives. If we did not work, we would all very quickly turn to violence and crime. To find out more about this I conducted a survey of 249 random subjects.
From this we can see that people are clearly divided on the topic. But I was asking about work in general. If, however you ask someone whether their job has meaning you get a very different response. This is demonstrated by a 2015 YouGov poll. The poll showed that 37% of people do not believe that their job is contributing to the world (Yougov, 2015). This is a shocking statistic and makes us wonder quite what jobs these people are doing that they their jobs as pointless.
By analysing the data in the survey, we can conclude that working class people are more likely to believe that their job is not having a meaningful contribution to the world than middle class people. We can also see that some areas have significantly lower levels of meaningfulness than others such as London. Despite the elevated levels of meaninglessness in London less people said that they would be “not proud to tell a stranger what their job was”, unlike in Scotland where there are elevated levels of meaningfulness and elevated levels of shame.
So, we can conclude that people in lower economic brackets are more likely to see themselves in pointless jobs. We can also see that people in areas of high population concentration such as London and the north are more likely to be not fulfilled by their job.
In August 2013 David Graeber wrote an influential article for STRIKE! Magazine (Graeber, 2013).In this article he argued that many modern jobs are ‘bullshit jobs’. They point out that in 1930 John Maynard Keynes (arguably the capitalist equivalent to Karl Marx), predicted that by the centuries end that developed countries such as Great Britain would be so technologically advanced that people who lived there would on average work only 15 hours a week. And yes, as predicted most of manufacturing jobs have been automated yet despite this we have not achieved the 15-hour week. Graeber argues that this is due to the creation of ‘bullshit jobs’. There has been a massive explosion in the services/administration sector. In fact, between 1948 and 2011 the services sector in the US has
gone from 45% of total employment to 68% of total employment (not including government jobs) (The Economist, 2014)
The new services sector comprises many jobs such as:
· Financial services
· Telemarketing
· Corporate law
· Academic/health administration
· Human resources
· Public relations
These are what Graeber proposes are ‘bullshit jobs’. A bullshit hob is one that provides little or no meaning to society and the world. And yet even though the people doing these jobs find them pointless they continue to do them. And what’s more they continue to be created.
If bullshit jobs are pointless why are they created? Many would argue that society creates jobs to ensure that they can continue to partake in society. Some would argue that because of this if people did not have to work to have a good enough income to live on then they would not work. They argue that instead they would spend their time doing things they enjoy and getting the education required to do interesting jobs such as medicine or teaching. This is backed up by universal basic income studies. A universal basic income is a guaranteed income that is paid to all eligible members of society. This is often done by a negative income tax; this is where after earning below a certain point the state stats to give you a guaranteed income. Most importantly however this payment has no strings attached. This means that if people want to then they can and can do no work and just live of benefits. However, the statistics from the studies do not show that this happens. In 1974 a basic income study was carried out in Manitoba (Canada); it showed that people barely reduced their working hours, and those that did used it to spend more time with their families and or taking additional classes reaping untold benefits for the economy (Hum & Simpson, 1993).
Many argue that even if automation does occur then people could continue to do jobs that give them meaning if they wish. Just because a job could be done by a robot does not necessarily mean it will be. If people find meaning in work, then they can continue to do so. However, if your job is mind bogglingly boring then why should you have to do it if you don’t want to? As we enter the new automated age then we are going to have to realize that we should have fun in life and if that means not working then so be it. But the clear majority will find something to do be it inventing, painting, or pushing the boundaries we must accept that our society will change to accommodate our new-found freedom.
How can we mitigate its impact?
Working on the dual assumptions that; soon robotic automation will increase so that 30% of jobs become automated (with not enough being created to replace them), and that in our current state if we get rid of work then there would be large increases in crime and violence. We can conclude that preemptive measures need to take preemptive measures to mitigate the impact. I have split these preemptive measures into two main types.
Only by combining a variety of government policies and regulation with a collective societal move towards less work based system we can ensure that minimal damage is done. This is the main subject of this report. I will first discuss potential government policies and then the action that society must implement to make the most of automation.
Government policies and responsibillities
Government policies come in the form of taxes, benefits, regulation, or programs. A tax is designed to incite a behavior using negative reinforcement i.e. persuade someone or a company to do something otherwise they will lose more money. Benefits give money to people (typically working-class people), this provides them with an income to survive even if they lose their jobs. Regulation prevents the development ‘bad robots’ such as terminators. Programs run by governments help to retrain people to get them new jobs by giving them new skills such as programming.
Tax
The tax I am investigate is a robot tax. A robot tax is a system where cooperation’s are taxed depending on how much of their workforce is automated. For instance, if you were a company that ‘employed’ a robot corporate lawyer you would pay robot tax equivalent to the income tax a corporate lawyer would have paid. This money could be used to fund other government initiatives such as new benefits and retraining programs (Varoufakis, 2017). Proponents for the tax are wide ranging and include tech giants such as Bill Gates (Gates, 2017) and futurists such as Elon Musk (Musk, 2016). However, some people such as Estonian politician Andrus Ansip believe that this is a bad idea (Ansip, 2017). It is argued that it would be difficult if not impossible to calculate the equivalent wage that the robot would have earned if a human where doing the same job. Furthermore, it is argued that this would reduce innovation as it would stop companies automating jobs, this is bad as some jobs a very dangerous and it is ethical to automate them even if it means someone loses their job (Isaac & Wallace, 2017).
Benifits
A common suggestion to mitigate the impact of robotics is the implementation of a new benefit called a Universal Basic Income (UBI), it is also known as basic income (BI), citizens income (CI) and negative income tax (NIT). But whatever its name (I shall use UBI) it involves giving all citizens a basic income (except in NIT where it is only the poorest) (Basic Income Earth Network, 2017). It has been studied in many studies in a range of situations for a variety of clients. It is argued that doing so would be cheaper than our current welfare system, this is because there would be very low administration costs. Furthermore, it is argued (and proven in studies) that a basic income gives better outcome than independent benefits (Hum & Simpson, 1993). It is also shown to increase personal development and entrepreneurship as people have a safety floor to stand on to achieve their aims be it setting up a company or training to get into a new profession. This is how UBI solves the issue of automation, it encourages personal retraining and entrepreneurship which in turn provides new jobs and bolsters the economy. Opponents argue that a UBI would encourage crime and antisocial behavior such as drug abuse. However, a report by the world bank that summarized the findings of 30 studies disproved this (Evans & Popova, 2014).
Regulation
One big worry about robots is that they will rise and take over the world. Whilst this may at first seem like an unrealistic and reactionary response to automation. However, these fears are well founded. In 2015 a robot was released by Queensland university of technology that will patrol coral reefs, and autonomously make the decision to kill the deadly crown of thorns starfish that destroys reefs (Dayoub, Dunbabin, & Corke, 2015). Although this application is undeniably good as we need to protect corals; it sets a dangerous precedent. The same technology can easily be expanded into military drones. Drones have long been used by the military, however this has led to sometimes disastrous consequences. The pilots feel detached and say it is like stepping on an ant (Pilkington, 2015). Imagine how much that feeling of detachment will become when instead of pulling a trigger you just must sign a piece of paper to authorize the strike. Despite this and warnings from high profile critics such as Stephen Hawking, Elon Musk, and Steve Wozniak (Future of Life Institute, 2015). As such it is undeniable that we should enact legislation to prevent the development of AI that decides when to kill human to ensure that we do not lose control.
programes
One proposed solution is retraining. This is where people who have been or will be made redundant due to automation are retrained to do new jobs. This retraining is funded by the government or previous employer and is usually in the form of a course or other qualification (Carson, 2015). These types of programs are useful and are a common way to mitigate impact when unemployment occurs on a mass scale. However, the type of unemployment that we will see might not end up being concentrated as it is normally. If all the manufacturing companies fired half their workers, yes there would be a lot of unemployment, but it would be widely dispersed; it is also harder to retrain people when they are dispersed as you cannot just set up one program. Therefore, these new courses will mostly have to be done online. But this again throws up another problem. The jobs that will be created/will not be automated are not manufacturing or laboring jobs, rather ones that require intelligence, independent/creative thinking, and human understanding (see next page) (McKinsey Global Institute, 2017). We can see that the jobs that will be automated the least are all degree level, education, management (less so with this one) and professionals. From this we can conclude that instead of providing standard retraining we need to other degree level retraining. To do this though the new students will have to pay tuition fees which are prohibitively high to some students let alone parents trying to support their own children going through uni who cannot access grants. In short if we want to mass retrain people at a degree level we need to get rid of tuition fees.
Societal action
Currently our society is geared to attain 100% employment. This full employment model creates pointless jobs just for the sake of keeping people working (Graeber, 2013). However, if 30% of people become unemployed this model will quickly fall apart. So undoubtedly retraining programs will appear and retraining some of the unemployed. But a large portion won’t want to be. If you are a lawyer, you’re not going to want to retrain into a teacher or a therapist because they’re completely different fields that wouldn’t interest you. And even if a UBI is implemented then we can’t all be entrepreneurs. This is mostly because it costs a lot less to run a successful company in the modern day. For instance, Instagram was bought for $1 billion, at that time it only had 13 employees (Geron, 2012). As this clearly shows you now need a lot less people to have an even bigger impact than ever before. So, we need to find something to occupy ourselves with.
Interplanatary colonisation
One suggestion is that we apply our newfound technological capabilities to undertaking a great task such as exploring space. This has several benefits.
1. It would retrain people
a. This is because starting a colony will take many new skills from all backgrounds. We could gear the retraining programs to train people to build rockets
2. It would produce employment
a. Yes, it might be much cheaper to build rockets by robot but why do that when you could employ people? On earth we could use the robots to do the mundane tasks that just have to be done such as; mass farming to feed everyone, building homes, treating illnesses.
3. Life would be less likely to be wiped out
a. We might just be the only life in the entire universe. Maybe even all of time. So it would be a real shame if we were wiped out by a single asteroid or a territorial spat or a massive plague. But if we have a self-sufficient colony on another world the chances of ALL of humanity drop to practically zero.
Despite these benefits there are some serous disadvantages. For instance we might accidently create a dystopia such as in Kim Stanley Robinsons Mars trilogy and 2312 (Robinson, The Complete Mars Trilogy: Red Mars, Green Mars, Blue Mars, 2015) (Robinson, 2312, 2013), if we want to avoid this we should ensure that the selection criteria for colonization is not financial but based on ability.
Elimination of the great killers
Throughout human history life has been short and nasty. If you were lucky enough to be born and your mother to have survived the ordeal you lived through roughly 40 grueling years of work to end up dead. By comparison even the poorest person in the first world would not suffer that much. However, many people in LIC’s (less industrialized countries) still live in this Malthusian misery trap. However, we now have the technological abilities to free them. We could use robots to mass farm to feed cheaply people (farmbot, 2018), we could use modified 3d printers with concrete to 3d print houses in areas with high homelessness (apis-cor, 2018), and we can release genetically engineered mosquitoes to crash the population of a certain type of mosquito (Carvalho DO, 2015). All these techniques use the latest in technology and robotics to solve the great problems of the world. However, to deploy we will need to work together with a large human fleet to support it.
A new social order
Automation itself will undoubtedly cause a great in politics. This is because as previously established society will have to change and so will our priorities. And seeming as political order and systems, I descended from those governed as per defined by social contract theory (Rousseau, 1913). However, as our society changes rapidly our systems will quickly unfold and become unsuitable for the modern world. This will inevitably lead to the creation of new types of government such as Futarchy (Buterin, 2014) and liquid democracy (Jochmann, 2012). However, if not properly handled the opportunity maybe seized by the ‘new radicals’ such as Donald Trump and Heinz-Christian Strache (Carswell, 2017). However, if we can seize the opportunity then we have a chance like no other to make a real lasting impact on the world.
Conclusion
Robotic automation will have a wide-ranging effect on society. The predicted levels of unemployment can only be described as catastrophic by today’s standards. To cope with this change, we must find meaning in our lives and our existence. To cope we will take on new and exciting challenges such as founding a Martian colony and becoming more than human. Sadly, though the governments that have the power to enact the decisions required to help humanity cope with the turbulence of change, seem blissfully ignorant of the dire need for discussion and debate on this most important debate.
Bibliography
Ansip, A. (2017, June 2). EU Commissioner Says No to Bill Gates’ Robot Tax Idea. (CNBC, Interviewer)
apis-cor. (2018, January 7). Home apis-cor. Retrieved from apis-cor: http://apis-cor.com/en
Arntz, M., Gregory, T., & Ulrich, Z. (2016). The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis. OECD Social, Employment and Migration Working. Paris: OECD Publishing. doi:http://dx.doi.org/10.1787/5jlz9h56dvq7-en
Basic Income Earth Network. (2017, December 28). BIEN: Basic Income Earth Network. Retrieved from About basic income: http://basicincome.org/basic-income/
BBC News technology. (2012, April 10). BBC. Retrieved from BBC|News|Technology|Facebook buys Instagram photo sharing network for $1bn: http://www.bbc.co.uk/news/technology-17658264
Berriman, R., & Hawksworth, J. (2017). Will robots steal our jobs? The potential impact of automation on the UK and other major economies. London: Price-waterhouse-Coopers LLP.
Buterin, V. (2014, August 21). An Introduction to Futarchy. Retrieved from Ethereum blog: https://blog.ethereum.org/2014/08/21/introduction-futarchy/
Carson, E. (2015, August 3). How workers can retrain for careers in an automated world. Retrieved from ZDnet: http://www.zdnet.com/article/how-workers-can-retrain-for-careers-in-an-automated-world/
Carvalho DO, M. A. (2015). Suppression of a Field Population of Aedes aegypti in Brazil by Sustained Release of Transgenic Male Mosquitoes. PLoS Negl Trop Dis, 1. Retrieved from https://doi.org/10.1371/journal.pntd.0003864
Center for Global Policy Solutions. (2017). Stick Shift: Autonomous Vehicles, Driving Jobs, and the Future of Work. Washington, DC: Center for Global Policy Solutions.
Coindesk. (2017, November 30). Price page, 2017–2018. Retrieved from Coindesk: https://www.coindesk.com/price/
Dayoub, F., Dunbabin, M., & Corke, P. (2015). Robotic Detection and Tracking of Crown-of-Thorns Starfish. Queensland: Queensland University of Technology.
Durden, T. (2017, Febuary 3). Rig Count Surges Again To 16-Month Highs (But Where’s The Oil Industry Jobs). Retrieved from ZeroHedge: http://www.zerohedge.com/news/2017-02-03/rig-count-surges-again-16-month-highs-wheres-oil-industry-jobs
Evans, D. K., & Popova, A. (2014). Cash transfers and remptation goods: a review of global evidence (English). Washington DC: World Bank. Retrieved from http://documents.worldbank.org/curated/en/617631468001808739/Cash-transfers-and-temptation-goods-a-review-of-global-evidence
farmbot. (2018, January 7). Home farmbot. Retrieved from Farmbot website: https://farm.bot/
Frey, C. B., & Osborne, M. A. (2013). THE FUTURE OF EMPLOYMENT: HOW SUSCEPTIBLE ARE JOBS TO COMPUTERISATION? Oxford: Oxford University.
Future of Life Institute. (2015, July 28). Autonomous Weapons: an Open Letter from AI & Robotics Researchers. Retrieved from Future of Life Institute: https://futureoflife.org/open-letter-autonomous-weapons/
Gates, B. (2017, Febuary 17). Why Bill Gates would tax robots. (Quartz, Interviewer)
Geron, T. (2012, September 6). Facebook Officially Closes Instagram Deal. Retrieved from Forbes: https://www.forbes.com/sites/tomiogeron/2012/09/06/facebook-officially-closes-instagram-deal/#6bed65c61d45
Graeber, D. (2013, August 1). On the Phenomenon of Bullshit Jobs: A Work Rant. Retrieved from STRIKE! Magazine: https://strikemag.org/bullshit-jobs
Gross, B. (2016). Culture Clash. Investment Outlook, 2. Retrieved from https://17eb94422c7de298ec1b-8601c126654e9663374c173ae837a562.ssl.cf1.rackcdn.com/Documents/umbrella%2Fbill%20gross%2FBill%20Gross%20Investment%20Outlook_May%202016.pdf
HM Treasury. (2017). Autumn Budget 2017. London: HM Treasury.
Hum, D., & Simpson, W. (1993). Economic Response to a Guaranteed Annual Income: Experience from Canada and the United States. Journal of Labor Economics, 11.
Isaac, A., & Wallace, T. (2017, September 27). Return of the Luddites: why a robot tax could never work. Retrieved from The Telegraph: www.telegraph.co.uk/business/2017/09/27/return-luddites-robot-tax-could-never-work/
Jochmann, J. (2012, November 18). Liquid Democracy In Simple Terms. Youtube. Retrieved January 7, 2018, from https://www.youtube.com/watch?v=fg0_Vhldz-8
Katz, L., & Verhage, J. (2017, November 27). Bloomberg Technology. Retrieved from Novogratz Says Crypto Will Be ‘Biggest Bubble of Our Lifetimes’: https://www.bloomberg.com/news/articles/2017-11-28/novogratz-says-bitcoin-to-win-out-over-other-digital-currencies
McKinsey Global Institute. (2017). A FUTURE THAT WORKS: AUTOMATION, EMPLOYMENT, AND PRODUCTIVITY. London: McKinsey&Company.
Musk, E. (2016, November 4). Elon Musk: Robots will take your jobs, government will have to pay your wage. (CNBC, Interviewer)
Pilkington, E. (2015, November 19). The Gaurdian. Retrieved from Life as a drone operator: ‘Ever step on ants and never give it another thought?’ : https://www.theguardian.com/world/2015/nov/18/life-as-a-drone-pilot-creech-air-force-base-nevada
Robinson, K. S. (2013). 2312. London: Orbit.
Robinson, K. S. (2015). The Complete Mars Trilogy: Red Mars, Green Mars, Blue Mars. New York City: Harper Voyager.
Rousseau, J. J. (1913). Social Contract & Discourses, Translated with Introduction by G. D. H. Cole. New York: Dutton&Co. Retrieved January 7, 2018, from http://www.bartleby.com/br/168.html
The Economist. (2014, Jannuary 18). The onrushing wave. Retrieved from The Economist: https://www.economist.com/news/briefing/21594264-previous-technological-innovation-has-always-delivered-more-long-run-employment-not-less
Varoufakis, Y. (2017, Febuary 27). A Tax on Robots? Retrieved from Project Syndicate: https://www.project-syndicate.org/commentary/bill-gates-tax-on-robots-by-yanis-varoufakis-2017-02?barrier=accessreg
Yougov. (2015, August 12). Yougov|News|37% of British workers think their jobs are meaningless. Retrieved from Yougov: https://yougov.co.uk/news/2015/08/12/british-jobs-meaningless/
X