Thoughts on the Tradeoff between technology growth and employment


As the developed countries of the world limp out of yet another crippling recession, macroeconomics has been forced to do some soul-searching. Graduating from the London School of Economics with a degree in Economics our departmental commencement speech, in front of friends, family and graduates alike, was a sombre one: impressing the need to manage the public’s expectations about what Economics and Economists are able to do. Put bluntly, despite all the great minds that have applied themselves for decades against the Economics grindstone the holy grail of creating a ‘no recession, no unemployment and steady growth economy’ has remained elusive.

In this essay I argue that the economies of the future are going to look vastly different from the ones we have today. In particular, I believe there is a trade-off between technology growth and full employment and that if we want the former we shall have to give up on the latter. We shall need to adapt to a society, and a politics, where most people, probably north of 90%, do not work at all in their entire lives. I argue that this is not only not a bad thing, but actually something we should work towards.


So first of all, let’s start with the assumption that for the foreseeable future, let’s say the next hundred years, we shall have strong technology and productivity growth. This is obviously a strong assumption and so we shall not neglect the alternative, that our future might be one with limited or perhaps even no technology growth, in fact this is something we shall examine in the second half of this essay. For now though, although perhaps optimistic, assuming technology growth is a useful place to start out our discussion.

My argument, which I have summarised in the table above, is that technology growth will lead to a net decrease in the ‘number of jobs’. The reason for this is simple: technology growth destroys jobs faster than it creates them.

This was not always the case, for example during the Industrial Revolution advances in farming equipment meant lots of farm workers were put out of jobs but this destruction of farming jobs was offset by the creation of lots of manufacturing jobs in the factories.

Now it is worth drawing a distinction between developed countries and developing countries. Developing countries in Africa or even countries like China and India could get away with doing nothing innovative for decades to come but still have fantastic ‘technology growth’ because they are simply imitating and executing ideas from the developed countries. This is most certainly worth doing and has resulted in massive improvements in material living standards for billions of people across the planet and hopefully will continue to do so. However, eventually these developing countries will catch up and become developed and then will face the same problem that we in the West are already facing now; which is to have sustained technology growth we need to keep coming up with new things.

Peter Thiel, co-founder of PayPal and billionaire investor in companies like Facebook, argues that the problem is there has been a failure of imagination. Thiel grew up on the science-fiction of the 50’s and 60’s which imagined a 21st century of underwater cities, colonies on Mars, as well as intergalactic trade and exploration. A world where we could extend lives into the hundreds of years, mind-share ideas and skills instantaneously and dare I say it teleport (‘Beam me up, Scotty!’). As the motto of his venture capital firm ‘Founders Fund’ wittily points out ‘We wanted flying cars, instead we got 140 characters.’

Although I agree that such projects, if possible, would be both very cool and worthwhile, I do not think they would create a lot of jobs. You only have to look at the most innovative companies of today like Google and Apple which employ workers, admittedly very high-skilled workers, in the 10,000s, not the millions, hardly sufficient to find jobs for the billions of people on the planet.

This is a problem because I believe there are three types of technology growth.

  1. Build new stuff e.g. iPhone
  2. Automation e.g. car manufacturing
  3. More productive employees e.g. computer

Both ‘automation’ and ‘more productive employees’ types of technology growth, as we shall see, both lower the number of jobs. Previously these forces were kept in balance by the first type, building new stuff but now when we build new stuff we don’t require lots of workers because most of the job creation is in the innovation of the product not in the manufacture or production of it, and for that you just need a few highly skilled workers. It is obvious why ‘automation,’ at least temporarily, destroys jobs because it involves the direct replacement of human labour with machines and robots. A classic example of this is the manufacture of cars where much of the production now is automated, whereas previously it was very labour intensive.

Why ‘more productive employees’ type of technology growth destroys jobs is more subtle. In economics, one of the key assumptions is unlimited demand with the only constraint being the budget constraint, i.e. not having enough money. With this assumption, even if people are unemployed new industries and products will develop to take advantage of these unused resources and to meet our insatiable demand. Of course, in the short-run there may be structural unemployment but in the long-run full employment is always possible, as with education and retraining workers are able to find work in the new industries being created. Now of course, for most of human history this is a fairly accurate picture and it is still probably an accurate picture for most people today, consider the Chinese rice farmer or the Brazilian favela housewife, I’m sure, if only they had the money, they would be buying fancy cars and going on luxury holidays. However, we are already seeing the effects of another constraint, a constraint that has usually gone unaccounted for and that is time. Limited time is easiest understood in the creative arts. Only so many people can earn a living as authors, or musicians or artists because you only have so much time to consume these things. And as everything is so scalable now why listen to the average pub singer in your local bar when you can listen to Adele on your iPod? This same principle extends to other areas outside of the creative arts. And as workers become more productive it will be possible to reach our ‘limited demands’ with fewer and fewer workers.

The question then is, faced with the reality of a world with fewer and fewer jobs, what should we do next? I would argue that rather than try and continue to pretend that economies with both high employment and technology growth are possible we should embrace the change. In fact, those same science-fiction novels that Thiel grew up on also imagined another change: that we would all be working single digit hour work weeks because robots had replaced all the menial tasks and allowed us to live lives of leisure. ‘The Jetsons’ a cartoon tv show in the 60’s that imagined a typical American family 100 years in the future (2062 to be exact) not only had a full-time robotic house-maid called Rosie but also had George Jenson, the patriarch of the family with a workweek, typical of his time, of just one hour a day, two days a week!

There is a practical aspect to such a society which is that there needs to be transference of ‘income’ from the robots, that are automating all the tasks that are currently being done by humans, to the humans who are now ‘unemployed.’ It is worth pointing out that the robots will not mind being paid nothing, because they are robots! This most likely would be done through a huge welfare state, where perhaps 90%+ of people would spend their entire lives without working. Now obviously, welfare states often get a bad reputation because there are always people who abuse the system and the unfairness of taxing those who work to support those who do not work. In particular, although the production of most things will be done by robots, the machines obviously do not collect a wage and so the actual transference of income will be from the entrepreneurs who made the robots to the masses of unemployed, and with taxes north of 95% I’m sure economists around the globe will be having fits over the lack of incentives.

There are a few counterarguments to this. To start with I absolutely recognise humanity’s need for inequality, something that is perhaps even greater than its need for equality. First of all in a world where everyone has a high standard of living and does not need to work, money will matter less and there will be other avenues for ‘inequality’ or less cynically ‘a diversity of identity’ to be expressed, like through hobbies and interests. Furthermore, in a world where most people do not work but a few highly skilled entrepreneurs, scientists and artists do, there would be arguably too much inequality rather than too little, so forgetting all the baggage we have about tax rates, there should still be sufficient incentives to motivate great projects, even if taxes are at 95%. It is worth noting that it is not the single mum working 3 jobs who is being taxed 95%, she is now living a life of leisure, but rather the Bill Gates’ and Warren Buffetts of the world. Some may argue that a life of leisure is not fulfilling, work provides meaning and fulfillment which sitting at home and watching TV can never do. This I absolutely agree with but I think it is worth noting that most people do not have jobs anything close to our ideal of fulfilling work. Even in America, the most developed of nations, most people do very menial jobs. As the education reformist John Taylor Gatto in his ‘The Underground History of American Education’ points out the top ten most common jobs in 2020 is predicted to be, according to the US Bureau of Labour Statistics:

  1. Retail salesperson
  2. Registered nurse
  3. Cashier
  4. General office clerk
  5. Truck driver
  6. Managers
  7. Janitors, cleaners, domestic servants
  8. Nurse aids, orderlies, general attendants
  9. Food counter workers
  10. Waiters

So given that most people do unfulfilling, low-skill work I think the best thing is if we as a society try to automate each of these jobs. Google’s self-driving cars might be a step in the right direction in terms of replacing Truck Drivers, or supermarkets having self-checkouts, a first step in automating Cashier work.

With time as all these jobs are automated, people, with a guaranteed high standard of living, will then be free to pursue not only lives of leisure but also have the chance to improve themselves and both discover and follow their passions. I know I certainly would take more risks pursuing my entrepreneurial dreams if I was not so scared of, essentially, getting a bad job. Peter Thiel argues that increasingly, higher education is not about learning anymore but rather about insurance, which says that if you have a degree certificate you will not, what he describes as ‘fall through the cracks in society.’ So rather than damage incentives there may be actually more innovation and more production. Just think of the number of Marlon Brandos who never tried to become actors because they were scared of spending their lives as waiters! Now as I argued previously most people will not be world class in anything, that by definition is rare but with the freedom to fail there is nothing wrong with that.


With the transition to a world where we have less than 2 children per woman population growth is no longer going to be an engine of economic growth anymore. This is a good thing as resource constraints would probably mean anything above this would end in a Malthusian catastrophe. Furthermore changing demographics, notably the boost to our economy that bringing women into the workforce has had has now largely run its course. Thus, in the developed world at least, the buck has officially been passed and further economic growth will require technology growth.

It could be argued that a world with no technological growth is not such a bad thing. After all material living standards were largely the same from the Roman times all the way through to the beginning of the Industrial Revolution, the last 200 years have been the exception rather than the rule. And furthermore, there is more to life than material things.

Nonetheless there are I think good reasons for wanting economic growth. Firstly it is nice not only for individual people but for humanity as a whole to progress. Also everything from our politics to our economics works better when we have it. As Peter Thiel points out, if we have no economic growth, i.e. the proverbial pie is not getting any bigger, then suddenly democracy, which is based on a spirit of win-win compromises, turns into a zero-sum world where for one party to do better another has to do worse, a breeding ground for societal division and even political extremism.

So given that we have established technology growth and the economic growth it brings is worth having the question then is who is going to give it to us? There are a number of candidates.


With limited technological growth our politics have become increasingly zero-sum. Given our transition to a welfare state as well as the high debt levels it is extremely difficult for governments to have the political capital and financial capital to invest in big technology projects. In the past, this was not the case with the government being instrumental in, for example, both the Manhattan Project and the Space Race. However, both of those programs were possible because there was an outside threat to motivate them (Japan and the Soviet Union respectively). I heard that both China and India are looking to ramp up their own respective Space programmes so it possible this could be a productive rivalry although the obvious downside is the threat of real war.

There is also the usual argument of creating ‘innovation friendly’ tax codes and laws. Peter Thiel has argued that areas like energy and medicine have suffered because of over regulation where the government has tried to prevent bad outcomes at the cost of eliminating the chance for good ones. There is a trade-off because innovation is messy and requires lots of experimentation, even something as scientific as science, and arguably we have become too risk averse. In contrast, the only two areas that were not regulated were the only two areas that experienced innovation: computers and finance, although with finance we perhaps got the balance too far the other way!


Universities are unique in that a lot of research that goes on has no immediate payoff. This is good for two reasons 1) there is more to life than creating technology growth and 2) a lot of the most valuable projects are those with no immediate payoff and therefore outside the scope of business. Even though I do not think it is possible for everyone to be employed in high skilled jobs (among other things because of the time constraint) clearly education is very important for future technological progress. I think one measure that would be useful would be to outsource some of the undergraduate lectures because not only might it improve the quality of teaching but more importantly it would also free up more time for professors, to do what they do best, which is to focus on their research (I have written a long article on this topic and I also have a little start-up project on this, contact me for details:


I think the fundamental problem with Economics is that is operates at the 2nd order, its’ about fine-tuning but does not have the power to make big changes to economies. Progress towards more and better economic theory will therefore also be of 2nd order importance. At the heart of Economics is the fundamental assumption that if Economists (largely through governments) do good economics, that creates the best possible conditions for growth then growth will happen. I would argue however, that most of the big macro variables we care about are ultimately out of the control of both our Governments and our Economics.


Start-ups are wonderful because they allow new ideas, products and methods to be tested with limited risk (from a society standpoint). Some of the most famous companies today including Disney, Amazon, Apple and Google were literally founded in garages. And there are several more that were founded at university or by drop-outs such as Facebook, Dell, Microsoft, Virgin, Oracle, McDonalds, Slok Group, Hutchinson Whampoa and General Electric. I certainly believe that much of the technology required to automate many of the ‘menial jobs’ we discussed above may come from start-ups as they are probably primarily robotics and software problems which are not too capital intensive.

Nonetheless many of the biggest and most important projects, particularly some of the more fanciful science-fiction projects, require investments into the hundreds of millions or even billions well beyond the scope of most start-ups with even the best possible venture funding. As an example, Elon Musk who made more than $300,000 million co-founding PayPal could not find sufficient outside investment to fund his rocket ship company Space X and his electric car company Tesla and instead had to invest all of his personal fortune to save the two companies in the Financial Crisis of 2008. Clearly if we require big projects to be tackled by billionaire entrepreneurs who are willing to risk all their personal fortune then we are greatly limiting our chances for technology growth.

Big Companies

So really our best bet should be big companies. Peter Thiel in an event on technology growth with Google’s Eric Schmidt argued that Google’s failure to invest the $50 billion or so it has sitting on its balance sheet proves that Google has insufficient vision to put that money to good work. Schmidt countered that Google has done lots of innovative projects and that there are other limits to innovation than a lack of money.

The most obvious of which are diseconomies of scale and the well documented failures of big companies to innovate. Robert Lutz of General Motors Company who led the development of Chevrolet’s electric car ‘the Volt’, after seeing the success of Tesla Motor’s electric car, said it was embarrassing that it took Elon Musk’s Tesla, a start-up car company in Silicon Valley to show what was possible before General Motors a company with billions of dollars in revenue to make the same leap. Especially as General Motors, with competition from the Koreans and the Japanese, had presumably very strong incentives to innovate. One problem I think Management as a discipline should look to solve is how to build firms that do not suffer from diseconomies of scale, whether the solution is the right incentives, the structure of the firm etc. Part of the problem I think is that academic research is often impenetrable, if for no other reason that the volume of research, to an ordinary entrepreneur, who with limited time already, is not going to be able to sift through to find relevant papers that could help him run his company better. One alternative firm structure that I think that might work is rather than the top-down approach to resource allocation which you have in most companies, where people play politics to get into positions of power and influence, you could instead have a venture capital model of resource allocation within companies where ‘managers’ would be pitched to by teams of employees who want investment for projects. These ‘start-ups within companies’ could then benefit from all the economies of scale of big companies without the downsides.

The lack of innovation from big companies may also be due to a lack of high skilled labour. This may be in part because genius is rare; it may also be an education problem. I do think another problem that needs to be solved is the importance of ‘culture’ and teamwork skills within firms. To put it bluntly a lot of ‘creative people’ are not ‘corporate people’ and so that genius is left untapped because unless they found a company themselves (like Steve Jobs) they do not get past the interview stage or climb the management ranks. One answer could be to systematically introduce teaching these soft skills into the education system. Another alternative might be to structure firms so the soft skills matter less.


If our future is one without technological growth then I think our best bet is to try and transition to an economy with technological growth, where we have big companies, start-ups and universities focusing on automating our list of ‘menial jobs.’

With this successfully achieved we can then transition to an economy and society where most people do not work, living off the transfers from the welfare state. Even if my proposition is only partially true that still might mean economies with 20 or 30% unemployment, numbers that would be intolerable today. There will still, of course, be scope for big technology projects as well as the more general progress in the arts, science and business although their high-skilled nature and ‘the time constraint’ will mean these will not be particularly labour intensive.

If Technology Growth does lead to unemployment then what then?

I wrote an article in which I argued that technology growth will lead to mass unemployment and sent it to my friends asking their opinion on it. Although it was a radical and very contrarian view no one thus far has been able to convince me that it’s obviously wrong (pending further feedback!). Therefore after a couple of months I suddenly thought it might be cool to try and think through what, if on the off-chance my narrative about technology destroying jobs is correct, what that would mean for the world and crucially what the required transition to that new form of society and economy would look like. I think the end equilibrium of a society with mass unemployment is actually quite an attractive one, my Dad even joked that I thought up this essay because I want to live in a world where I don’t have to work! However, the more I think about the transition the more I think it will be fraught with difficulty. Ideally I’d want to be able to outline a blow by blow account of what I expect to happen but the big issue I’m struggling with is the sequencing of events. It would be cool if there was some framework to help think through the issues but I don’t know of any. So, instead in this essay I’ll try and outline a few dimensions that I think are worth considering.


I think if the Government still maintains the opinion that the country is capable of continuing as a low unemployment economy then there is a real danger of misguided Government Policy. In an interview with Charlie Rose legendary fund manager Jeremy Grantham argues that even now the government is too optimistic:

Jeremy Grantham: ‘I think Bernanke is whipping this donkey that can only grow at 1%, because he thinks that it’s a race-horse that should be growing at 3. So he’s gonna keep whipping this donkey. Charlie Rose: ‘This donkey can’t run.’ Jeremy Grantham: ‘Until it either drops dead or turns into a racehorse.’ Charlie Rose: ‘And you’re betting on dead.’

Accordingly if the Government expects low unemployment it may in an effort to stimulate the economy print too much money leading to inflation or go into unmanageable debt trying to stimulate the economy. Perhaps dangerous is the fact that with high unemployment there could be great political instability and perhaps even the potential for extremism to take hold as we saw in 1930’s Germany after the Great Depression. If my assertion is correct it is therefore vital that everyone is educated and persuaded that a low unemployment economy is no longer possible.


Although I suspect the long-run equilibrium of high unemployment might be fairly stable with those who work only being those who enjoy it and/or earn extremely high returns for their work in the short and medium term there is potential for huge instability.

I think a central problem is that even if most menial jobs eventually get automated this will be a gradual process. As an example, I heard that efforts are being made to create a gate like the one that you walk through at airport security which as you push your trolley through scans all your items and bills you automatically. So let’s assume this technology works and therefore the cashier job is one of the first to go. Suddenly you have a lot of unemployed people. Can you start giving them the high standards of living that would be possible in a full roboticised society? Well if you did increase the welfare benefits to a comfortable level then suddenly all those people working in the undesirable, low paid jobs that have yet to be automated, let’s say cleaning jobs would choose not to work. Leading to either much higher wages for these undesirable jobs or perhaps those jobs just not getting done. Alternatively if you keep welfare at a low level then I think as unemployment year on year increases you will get more and more unrest potentially leading to the political instability and extremism already cited.

I think one idea would to help foster a more peaceful transition would be to have education extend later and later into life and retirement to occur earlier and earlier. This gradual erosion of the number of working years might act as a escape valve for all that unemployment pressure.


However, I think even the long-run equilibrium is fraught with a major problem which is international competition over tax revenues, i.e. the tax paying companies and workers. Countries in a bid to attract the limited number of companies and workers lower taxes because if the tax base is very small then welfare benefits have to be lowered potentially leading to civil unrest from the unemployed. Even if successful in attracting said companies and workers it comes with a cost which how did you get those workers and companies? Well lowered taxes and therefore lower welfare benefits and potentially the same problem as before of civil unrest among the unemployed.

Even if international cooperation and taxes could be agreed upon there are other factors that lead to the choices of where companies and people choose to locate. As it is impossble to have a completely equal distribution of companies and people across different countries this could lead to conflict. The only solution I can imagine is for a world wide tax and welfare system which would require such radical changes in our government and societies it’s hard to see it happening smoothly, if at all.

My first exposure to Economics came in the afterglow of 10 years without recession and discussion in the UK that Gordon Brown had finally beaten the boom and bust cycle. However, just at the moment I started studying Economics the greatest Depression since the Great Depression happened so inevitably I, like many of my generation, am on the look out for other assumptions that are perhaps unfounded. Although I’d hate to be pessimistic I think it is an interesting thought experiment to wonder what it would take for two major powers to go to war with each other. I suspect differences in ideology is unlikely to be a sufficient criteria (especially if one of the parties is a democracy) but economic unrest caused by mass unemployment and international migration of the best workers and companies seems unfortunately very plausible.


Finally, I think there is a question about the nature of identity and a good life in a world where most may never work. I think given the previously mentioned issue this is a relatively minor issue and ideas about how to have a fulfilling life are ones that I’ve explored in my previously cited technology leads to mass unemployment essay.