Technology’s Socio-Economic Problem
Some of us read about how self-driving trucks has resulted in unions pushing back over the loss of trucking jobs. We routinely look up “automation” and fear that our job may become that of a robot someday.
There exists a more insidious problem — the inability to catch an ever-increasing gap.
Take for example a vertically integrated company that runs an e-commerce and logistics business. We can represent some of the jobs such a company has as a job distribution:

On one hand, there are jobs which are extremely easy to automate. Robots are able to replace packers easily. On the other hand, the critical thinking skills an operations research analyst requires to chart the strategic direction of such a company is not easily replaceable. However, the diversity of skills sets is prevalent. In such a company, there needs to be people who are brawny to be packers, people with good IT skills to become developers and good people skills to be perform door-to-door advertising and sales.
Let us suppose the company embarks on an automation and digital campaign to cut costs and optimise the business. There will be new jobs created, and old ones would disappear. As part of an automation digital campaign, it typically entails:
- substitution of error-prone manual labour for robots with greater accuracy
- increased online presence
- the need to optimise robots, analytics algorithms and data visualisation tools to make more clinical decisions.
This skews the job distribution.

In such an automation campaign, the company might test out self-driven trucks, which will reduce driver demand. Packers will be almost obsoleted as the packing process automated by robots which will know how to load trucks. Salespeople will not be as needed as platforms become more digital.
On the other hand, with more digital content, there will be greater demand for hardware and software engineers. The company may want to perform further optimisation, further increasing the need for analysts. We have, through this example, created a structural unemployment condition.
You might ask, “If we train them, can they take on these new jobs?”
In theory, yes. In practice, it is far more difficult to train competent staff, let alone great ones. The reason is because of an overall drop in skillset diversity. Manual labour is eschewed for digital skills. Repetitive tasks are substituted for tasks requiring critical thinking and on-the-fly decision-making. With more automation, data collection becomes far more effective. The future worker must be competent with understanding data acquisition, manipulation and data-driven decision-making. All of these point to the need to think algorithmically. Not everyone is able to be armed with such a skillset this easily.
For those that cannot keep up, they will be condemned to rotating among a slowly decreasing job market. The squeeze and urge to retrain will be more pressing, but retraining does not happen on-demand.
As digitalisation and automation continue, even more data-related jobs will be created. However, even more manual jobs will disappear. The high demand for data-related jobs will lead to wage increases in those domains, especially with a labour shortage. Meanwhile, manual jobs will continue to have their wages depressed. As structural unemployment becomes more serious, society runs a risk of a permanent underclass that cannot adapt to the pace of change.
Such a technology-driven ecosystem shakes up society. There will be the rich, upper class, that own and exploit technology for their political and business objectives, the middle class that can operate technology to sustain a family, and an underclass that cannot keep up. This results in a new form of inequality: a knowledge-based inequality that centres itself on accessibility and knowledge of the technological world.
The owners of technology business are cognizant of the dangers of class struggles and unrest. The underclass may never catch up with the frenetic pace of the technological world. However, jobless people can be dangerous. If they had nothing to lose, they might take to the streets, commit crime, or engage in acts of sabotage; all of these hurt businesses. It is not entirely implausible that business owners might think of an attempt to “pay off” such troubles. This, rephrased, is one interpretation of the universal basic income (UBI).
UBI may not be entirely without merit. For one, it keeps citizens of a society that can never catch up with progress a living wage and a chance to survive. Politicians flirt with the idea because free money can result in votes for their continued careers in electoral office. Businesses flirt with the idea because it aids in establishing civil order. However, critics suggest that UBI could also be the igniter of a class warfare between the net givers and net receivers. UBI is nothing more than another method of social welfare transfer.
Much of these discussions about the tech underclass might be unchartered territory. Many issues in the technology space are unprecedented. For the first time in history, a private company (Facebook) has greater influence than some nation-states. For the first time, technology can cause side-effects in such short timescales the world has never experienced before. The pace in which we are changing is nothing short of a revolution much quicker than the first three.
The uncertain, non-linear world provides scope for us to shape our world. However, we must be aware that the extreme speed of change will leave people behind in its wake. Not everyone can think algorithmically. Not everyone is able to exploit technology to their advantage. Not everyone understands the data-driven world. All of us, however, can understand that human emotions, perceptions and psychology cannot be ignored in a world that is very much still about human society, and people privileged to understand technology need to understand the socio-economic implications of technology.