The Last Bastion of Human Employment
By Andrea Zorzetto & Christine Yong
What is the backbone of a big corporation like Amazon or Facebook? Jeff Bezos and Mark Zuckerberg, their founders and CEOs, inevitably come to mind. As the labour market is radically transformed by emerging technology over the coming decades, we can imagine a world where the management of others is among the last remaining preserves of human employment, while other tasks are automated or outsourced. So far, disruption has occurred within individual industries, but technological change is set to redefine the very structure of organisations, as the Fordist revolution did about a century ago. The new corporation will look a lot like Google, with a highly-skilled workforce, tiny compared to its annual revenue, and much of the value created by algorithms or outsourced to its users.
Two considerations are important in speculating about the future of work. First, jobs often involve a mosaic of different tasks, each requiring different capabilities. Consultancy McKinsey states that while fewer than 5% of occupations can be entirely automated now, about 60% of occupations could have almost a third of their constituent activities automated. Even a CEO is likely to see around 20% of its functions automated. Tasks are thus more appropriate elements of analysis than jobs. Second, we are after all human, and diverse in our desires and values. Developments in technology can abruptly change which tasks can be automated, but other factors too are at play: elderly care robots would be more acceptable in Japan, where the workforce is shrinking and people are receptive to automation, than in Brazil, where labour costs less.
Given these considerations, what changes can we expect to see in tasks across different sectors and occupations?
Artificial intelligence (AI) is proving a formidable force. Where mechanical automation has for decades substituted for many routine physical tasks, deep learning now threatens to automate even complex cognitive problems. As the efficiency of artificial neural networks to mimic humanlevel skills increases, we might expect algorithms to replace predictable cognitive tasks such as data collection and processing. Further in the future, even less routine tasks like stakeholder interactions and applying expertise are at risk. Even physicians will see their tasks evolve; IBM’s Watson is already capable of parsing research to suggest treatments to patients. However, a large proportion of tasks performed by a physician, and in many other lines of work, retains some unpredictability in the physical, cognitive and emotional domains. The technology to viably automate staff management, teaching or weeding a garden with algorithmic logic are, for now, beyond the horizon.
In the near future, however, many activities are more likely to be outsourced rather than automated, becoming part of what is most commonly known as the “gig economy”. The rise of ICT allows quick matching of supply and demand without intermediaries. Companies like Uber and AirBnB have undergone explosive growth in value. They connect self-employed producers with casual consumers in a heartbeat, generating economic opportunities and bypassing traditional costs in time and flexibility. Its pervasiveness in the public sphere is evident in the emerging lifestyle tropes of digital nomads, professional freelancers and influencers. Jobs comprising unpredictable tasks that cannot yet be viably automated are especially vulnerable to these “disruptive” forces. The idea is not novel; the design industry has relied on freelancers for decades. Now, we have the technology to directly and massively engage with each other, for tasks as meaningful as teaching or trivial as weeding the garden. Should we then expect the notion of the classical firm to crumble? In Coase’s definition, the firm exists simply to internalise and reduce the cost of transactions. This is now approaching negligibility, thanks to online freelance and classifieds platforms such as Upwork and Craigslist, respectively.
In the wake of these changes, the one core activity that will remain integral to traditional firms is management, for two key reasons. First, outsourcing management would dramatically worsen the principal-agent problem. Second, there is low risk of automation faced by management tasks. Conventional firms will become so lean that management remains as a bastion of human employment.
How should policymakers react to this tremendous disruption of the nature of work, caused by the combined rise of AI and the gig economy? Much of the public debate focuses on the difficulty of imagining a low-work society, in which redistribution would be crucial. However, it is uncertain whether the current technological revolution will be different from previous ones, which on balance created new and better jobs. Conversely, the explosion in the number of freelancers is a clear trend, and this requires to radically transform our tax and benefit systems. At any rate, it is evident that our institutions and perspectives will need to evolve with the times, and not against them.
Bibliography
1. Thompson, Derek. “How Air-Conditioning Invented the Modern World.” The Atlantic. September 19, 2017. Accessed October 09, 2017. https://www.theatlantic.com/business/archive/2017/09/tim-harford-50-inventions/540276/.
2. Chui, Michael, James Manyika, and Mehdi Miremadi. “Where machines could replace humans — and where they can’t (yet).” McKinsey Quarterly, July 2016. Accessed October 9, 2017. https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/where-machines-could-replace-humans-and-where-they-cant-yet.
3. Chui, Michael, James Manyika, and Mehdi Miremadi. “Four fundamentals of workplace automation.” McKinsey Quarterly, November 2015. Accessed October 9, 2017. https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/four-fundamentals-of-workplace-automation.
