Automation does not simply destroy jobs; historically it destroys some and creates others, while raising productivity. The real economic problem is the transition: when technology changes which jobs exist, workers whose skills no longer match what the economy demands face structural unemployment, and that adjustment is slow and painful for the people caught in it. The policy responses each carry a trade-off. Retraining and skills upgrading address the root cause but are slow and imperfect. A minimum wage protects low-wage workers, but a minimum set above the market-clearing wage can create a surplus of labour, that is, unemployment. A universal basic income provides a simple income floor, but it is costly and raises questions about funding and work incentives. The strong A level answer weighs these against one another rather than predicting the future.
Watch the technology stories for long enough and a pattern repeats. Cars that steer themselves inch out of the test track and onto real roads. Trucks run trial routes with no one in the cab. Parcels arrive by drone in a pilot scheme somewhere. Each headline lands with the same anxious question underneath it, and it is worth stating plainly: if a machine can drive the lorry, deliver the meal and ferry the passenger, what becomes of the people who do those jobs today? Driving and delivery are not a niche. They are a large class of work, and a great many livelihoods sit inside it.
I want to take that question seriously, because it deserves better than either of the easy answers. One easy answer is panic, the machines are coming for all of us and mass unemployment is inevitable. The other is breezy denial, technology has always created more jobs than it destroyed, so relax. The truth is more demanding and, frankly, more interesting, and the A level syllabus hands you exactly the apparatus to think it through. This is an essay about the labour market, and it is one of the clearest live cases you will meet.
The economics: structural unemployment
Start with the right label. When a technology changes which jobs exist, the workers it displaces are not lazy and the economy is not simply short of demand. The problem is a mismatch: the skills these workers have no longer match the skills the economy now wants. That is structural unemployment, and it is the precise concept the automation debate is about. A driver whose route is automated does not instantly become a software technician; the gap between the old skill and the new one is the whole of the difficulty.
- Structural unemployment
- Unemployment that arises when the skills workers have no longer match the skills the economy demands, often because technology has changed which jobs exist. It is a mismatch, not a shortage of total spending.
- Labour market adjustment
- The process by which displaced workers move, or fail to move, into the new jobs an economy creates. It can be slow, because retraining, relocating and rebuilding experience all take time.
- Productivity
- Output per worker. Automation tends to raise it, which is the same force that makes new goods, services and jobs possible, even as it removes the old ones.
Hold on to two facts at once, because the strong answer needs both. The first is that automation is genuinely disruptive: the jobs it removes are real jobs held by real people, and the labour market does not re-clear overnight. The second is that automation also raises productivity, lowers costs and, across the sweep of history, has created new kinds of work that did not exist before, from the roles around the motor car to the entire digital economy. Both are true. The mistake is to grab one and drop the other.
So the economic problem is not really a prediction about whether the robots win. It is the transition. Even if the economy eventually creates as many jobs as it loses, the displaced worker does not live in the long run; they live in the months and years it takes to retrain, find new work and rebuild a wage. That gap, painful and concentrated on particular people in particular trades, is what policy is actually trying to address.
Mr Toh's take
Here is what I find myself thinking about, and I put it to my students as a set of open questions rather than a forecast. We will, before very long, have to decide seriously what we do with drivers. When cars, trucks, food delivery and parcel delivery go fully driverless, or move to drones, that is not a handful of jobs at the margin; it is a large and visible class of work. I do not say this to frighten anyone. I say it because it is the kind of structural change an economist is supposed to look at squarely rather than wish away.
And once you take it seriously, you are pushed towards the bigger policy ideas, the ones that sound radical until the problem in front of you is large enough to need them. I think we have to be willing to discuss universal basic income, an income floor paid to everyone, and we have to think carefully about minimum wages, what they protect and what they cost. I am not telling you which of these is right; I genuinely think reasonable people land in different places. What I am sure of is that these are the questions, and a student who can lay them out as trade-offs, rather than as a slogan, is already thinking like an economist.
My own honest position is the unglamorous middle. The history of automation is that it destroys some jobs and creates others, so I do not believe the doom. But the transition is real, and it falls hardest on the people least able to absorb it, and that is precisely the policy problem. The interesting question is not whether automation is good or bad in the abstract. It is how a society carries the displaced across the gap, and at what cost, while it waits for the new jobs to arrive.
Treat automation as a structural unemployment question, not a vague essay about robots. Name the concept, explain the skills mismatch, then evaluate the policy responses against one another rather than predicting the future. A model sentence: "Because automation displaces workers whose skills no longer match those the economy demands, it raises structural rather than cyclical unemployment, so the appropriate response is supply side retraining to close the skills gap; a minimum wage or a universal basic income may protect incomes during the transition, but each carries a trade-off, and the stronger judgement weighs them rather than forecasting the level of future employment."
The policy responses, weighed against one another
This is where an evaluation question is really decided, so set the options side by side and be honest about the cost of each. None of them is a clean win, which is exactly why the comparison, rather than a single confident recommendation, is what earns the marks.
| Policy | What it does | Strength | Trade-off |
|---|---|---|---|
| Retraining and reskilling | Helps displaced workers gain the skills the new jobs demand | Addresses the root cause, the skills mismatch itself | Slow and imperfect; not everyone retrains successfully, and the gap is felt now |
| Minimum wage | Sets a legal floor under the wage of low-paid workers | Protects the incomes of the most vulnerable workers | If set above the market-clearing wage, a binding minimum can create a surplus of labour, that is, unemployment |
| Universal basic income | Pays every citizen an unconditional income floor | Simple, universal, and a cushion through the transition | Costly to fund, and it raises questions about work incentives and how it is paid for |
Three responses to technological unemployment, each with a real cost. The exam reward is in weighing them against one another, not in declaring one the answer.
Read the table as a set of tensions, not a menu. Retraining is the only option that actually closes the mismatch at its source, which is why economists reach for supply side skills policy first, but it is slow and it does not work for everyone, and the displaced worker needs to eat in the meantime. A minimum wage protects the low paid, yet a minimum pushed above the market-clearing wage can price some workers out and add to the very unemployment we are worried about. A universal basic income is clean and universal, but someone has to fund it, and we do not yet know how it changes the incentive to work. The honest conclusion is that the best mix depends on how fast the change comes and how well retraining holds up, which is a judgement, not a prophecy.
Automation destroys some jobs and creates others. The policy problem is the gap in between, and the people standing in it.
- Automation is a structural unemployment story. The issue is a skills mismatch, workers whose skills no longer match what the economy demands, not a shortage of total spending.
- It cuts both ways. Automation destroys some jobs and raises productivity, and historically it creates new kinds of work; both facts belong in a strong answer.
- The real problem is the transition, the slow, painful gap the displaced worker lives through while the new jobs arrive, and policy is mostly about carrying people across it.
- Each response has a trade-off. Retraining addresses the cause but is slow; a minimum wage protects workers but, if binding above the market wage, can cause unemployment; a UBI gives a floor but is costly and raises incentive questions.
- Evaluate, do not predict. The marks are in weighing the policies against one another for a given context, not in forecasting whether automation will cause mass unemployment.
Want this on paper? Grab the free 112 page Summary and Diagrams pack.
Frequently asked
Will automation cause unemployment?
Automation does not automatically cause lasting mass unemployment, but it does cause structural unemployment during the transition. When technology changes which jobs exist, workers whose skills no longer match what the economy demands are displaced, and re-employing them is slow because retraining, relocating and rebuilding experience all take time. Historically, automation has destroyed some jobs while creating others and raising productivity, so the long run picture is not one of ever-rising unemployment. The honest economic concern is not the long run total but the transition: the real, concentrated pain felt by the particular workers caught in the gap before the new jobs arrive, which is what policy tries to address.
What is structural unemployment?
Structural unemployment is unemployment that arises when the skills workers have no longer match the skills the economy demands, often because technology or the structure of the economy has changed which jobs exist. It is different from cyclical unemployment, which is caused by a fall in aggregate demand, and from frictional unemployment, the short gap while people move between jobs. The defining feature of structural unemployment is a mismatch rather than a shortage of total spending, which is why the usual remedy is supply side, retraining and reskilling to close the gap, rather than simply boosting demand. Automation is one of the clearest causes of it.
What is universal basic income?
A universal basic income, or UBI, is an unconditional payment made regularly to every citizen, regardless of whether they work, intended as an income floor everyone can rely on. Its appeal is simplicity and universality: it provides a cushion through disruptions such as automation without the complex conditions of targeted welfare. Its difficulties are equally real. It is expensive to fund across an entire population, which raises hard questions about taxation, and economists debate how an unconditional income would affect the incentive to work. For A level purposes, treat a UBI as one policy response to technological unemployment with a genuine trade-off, to be weighed against retraining and the minimum wage rather than accepted or dismissed outright.
Does a minimum wage cause unemployment?
A minimum wage can cause unemployment, but it depends on where it is set. If a binding minimum wage is set above the market-clearing wage, it raises the quantity of labour supplied and lowers the quantity demanded, creating a surplus of labour, which is unemployment, among the low-paid workers it is meant to help. If it is set at or below the market-clearing wage, it has little effect on employment. There are also arguments that in some labour markets a moderate minimum wage need not raise unemployment much. The exam point is to show the surplus on a labour market diagram and then evaluate: a minimum wage protects the incomes of low-wage workers, but a poorly set one can add to unemployment, so the effect is not automatic.
Is automation good or bad for the economy?
Automation is neither simply good nor simply bad; it is a structural change with real benefits and real costs that fall on different people. It raises productivity, lowers costs, and over history has created new kinds of work, which makes economies richer in the long run. At the same time it displaces workers in the jobs it removes, causing structural unemployment that is slow and painful to resolve, concentrated on particular trades and people. The economically serious view is to hold both sides at once and focus on the transition: how a society helps the displaced move into new work, and at what cost, through policies such as retraining, the minimum wage and a universal basic income, each with its own trade-off.
Worst subject to best
"Economics went from my worst subject to my best subject."
Gabrielle Goh ACJCThe standard, every week.
One essay or case study a week, personally marked with a worked model and a video walkthrough, from materials written by the author of the H1 and H2 TYS answer keys sold at Popular. This is the core JC1 and JC2 programme.
Weekly, marked, everything included
- A marked essay or case study each week
- Worked model plus a video walkthrough
- Onsite, live Zoom or recordings