I read an interesting fact a few months ago:
In 2012, Google, for example, generated a profit of nearly $14 billion while employing fewer than 38,000 people. Contrast that with the automotive industry. At peak employment in 1979, General Motors alone had nearly 840,000 workers but earned only about $11 billion—20 percent less than what Google raked in. And, yes, that’s after adjusting for inflation. – Rise of the Robots
It outlined a trend I’ve been well aware of for quite a while now which is the growing rise of machine automation. At first thought when reading the above most people see it as great news, Google was able to make 20% more profit whilst employing only 4.5% of the people GM did, that’s quite the achievement and they’re right! However it also bypasses a critical issue which is that it now takes considerably fewer people to do the same job.
I’m sure this isn’t shocking news to most people, there’s articles all over the web writing about it but I would like to address the issue from a more personal viewpoint. Most pieces just prattle on about this looming death cloud of robots “stealing our jobs” but what does that really mean for you and I? And more importantly what aspects are within our control and how can we prepare for and embrace this change?
The Evil Robots
Many seem to take “robots stealing our jobs” quite literally. They picture a humanoid shaped robot walking up to their desk, kicking their weak fleshy ass to the floor and telling them to beat it, the robots are running things now! And while some people believe that this WILL actually happen, most just ignore the whole argument because they’ve never seen a literal robot do such a thing. Even in advanced manufacturing plants humanoid robots basically don’t exist (it’s usually just a big arm), the closest thing I’ve ever seen is the Boston Dynamics Atlas robot (which is insanely freaking cool BTW).
So when all these forecasters and analysts and people predicting the future say that “robots will steal your job” people assume they mean literal humanoid robots… which don’t really exist… so their argument must be stupid and wrong.
But they’re not unfortunately.
Robots aren’t just going to steal our jobs in the future… they’ve already been doing it for decades. Haven’t you seen them?
You probably have. You likely see them every day and yet don’t even know it because “robots” in this context is mainly just a term for any computational automation which most commonly is done through software. You see, robots aren’t going to come in, boot you off your chair and sit down and start typing on your PC. No, that would be hugely inefficient and costly to make! It’s far, far cheaper and better to just have software do the task on that same PC you work on!
Businesses welcome this with open arms usually under the guise of “efficiency improvements” or “optimisation”. To be honest, I love it too! I constantly search for the most efficient, easiest way to do something and hell, one of the MAIN pillars of the MTM Strategy is to “Automate Your Repayments”! Most people in general love automation as it makes their life easier, usually cheaper and faster too. Some also love it because it means they don’t have to “deal” with people. My favourite example of this is the “self-checkout” systems that have been introduced into large food shopping stores. There used to be 15 isles of conveyor belts with 15 employees… now there’s maybe one or two of them and one of those is supervising the 15 automatic self-checkout systems.
This is what “robots stealing our jobs” TRULY looks like!
Studies into this have actually revealed that it is more pronounced when recessions occur. In the past the stock market would crash and burn and 10,000 jobs would be lost due to businesses contracting… but after some years all those jobs would be available again. Now instead the same crash occurs, the same 10,000 jobs are lost… but after some years only 1,000 jobs are available to do the work as automation has taken up the slack. The end result is that after each recession it now takes far longer for those jobs lost to be recovered.
In 2011 Apple built a massive $1 billion data centre that it uses to run its cloud services. Stretching across hundreds of acres, the now 5 year old facility only requires 50 employees to run.
Over time, as markets crash and are rebuilt or new businesses emerge new systems are coming online that more and more don’t need humans to operate and we as consumers of these services PREFER them. They are cheaper, faster, available 24/7 usually and provide better service in general (besides the dreaded “press 1 to blah blah blah” calls). So you have a perfect storm where businesses want to push forwards for obvious profit reasons and at the same time consumers are also constantly demanding these new features, cheaper prices and more availability too.
Virtually the only thing stopping all jobs being overwritten is that the technology isn’t there… yet.
A Robot Could NEVER Do My Job
Said the lift conductor. Said the manufacturer worker. Said the accountant. Said the lawyer. Said the doctor.
Another common way robots are stealing our jobs is by simply making them obsolete. Snail mail is a perfect example of this and AusPost has been having huge issues dealing with it. The manufacturers of records, tapes, CD’s and DVD’s are all growing largely irrelevant now that most get their media through online streaming. People correctly assume that it’d be all but “impossible” for a machine to build a DVD from start to finish with zero human input and they’re right! But now thanks to Netflix they’re unneeded.
Up until the last decade or so most of the job replacement has been in the physical domain. Robot arms to build everything from cars to phones, robot check-outs to let you buy your own groceries, online stores to eliminate physical stores, Airbnb to make the idea of a Hotel a bit antiquated or smartphones to make calculators, film, and more obsolete. However with the increase in computational power developed over the years and now the progress of machine learning it is breaching into the realm of white collar jobs more and more.
Automatic programs are now writing a huge number of online articles and you’ve probably already read them and not even noticed. Ironically, you can read about this through articles written online:
The Associated Press uses Automated Insights’ Wordsmith platform to create more than 3,000 financial reports per quarter. It published a story on Apple’s latest record-busting earnings within minutes of their release. Forbes uses Narrative Science’s Quill platform for similar efforts and refers to the firm as a partner.
Yes, that is major online newspapers printing “articles” that are entirely written by machine learning systems and few, if any, humans can tell the difference.
Watson’s successful diagnosis rate for lung cancer is 90 percent, compared to 50 percent for human doctors.
Doctors are a particularly interesting case because most people will tell you that they’d NEVER use a robot doctor over a human one. But ask yourself, how willing would you be to allow a computer to review your biopsy results and make the call on whether you have lung cancer given the above percentages? I’m sorry, but I’d pick the program every time.
Lawyers aren’t safe either, especially legal aides. The entire stock market is all but run by computers now and Accountants haven’t been safe for years. I’m sure you all know about automation in call centres and now even the managers are starting to feel the pinch too. Some of the biggest parts of being a (mid layer) manager is in collecting, formatting and reviewing company data and presenting that reviewed data (with recommendations) to their managers.
As our companies begin to employ more and more advanced systems that review, summarise and make predictions or recommendations there will be a smaller need for the human. Eventually companies will flatten their structures out, resulting in a few managers at the top and all the workers at the bottom with powerful automation and analysis tools constantly keeping track of metrics and issuing the most finely tuned industry standard (and profitable) recommendations to the top brass.
In a very recent publication Google announced that by handing the keys to their entire data centre cooling system over to a Deep Neural Net, managing over 80 monitored parameters and optimising them all minute by minute they saw a unheard of 40% decrease in energy used for cooling.
It’s painfully clear that automation is coming, to every job, to every field and the speed is increasing.
This is all very fascinating and fun to read about and watch but at the end of the day, how is this all going to affect YOU?
You don’t care if Dr Snooty Pants loses his fancy $500,000/year doctor job to the new and improved Dr Watson… you want to know what it will mean for your work, your job and your future. I get that. So I found you a really cool and handy online calculator that will tell you the likely hood of your job being automated (there’s also this one too) :-)
So now you have a bit more of an idea how susceptible your job is to automation over the next 10-20 years. Then what?
Well for me I realised long ago that this trend is not only already happening but will likely one day either see me out of a job or at the very least, make it incredibly hard to find and maintain a job, probably resulting in less and less pay. So rather than cry about it I decided to take matters into my own hand and put a priority on making sure that by the time 10-15 years is up that I won’t NEED a job.
The first step in this plan was obviously getting out of debt which meant paying off our mortgage. Fast. This has been one of the main drivers behind our relentless pursuit of becoming mortgage free. By the time it hits 2026, ten years from now, things are going to be significantly different in terms of finding and maintaining a job. That is a scary thing but it is also one of the most logical arguments I’ve found to make sure you’re in the best possible financial shape when it hits.
If you’re a legal aid and have been finding it harder and harder to find work eventually you’re going to have to leave that profession and re-train up on something else (preferably something that isn’t easy to automate). Wouldn’t that difficult life change be made simpler and easier without a mortgage debt hanging over your head?
Next week I’ll be discussing how paying off your mortgage also helps to set you up for the second part of dealing with this very significant threat of global automation. In the meantime though if you’ve looked up your profession and it’s a high percentage have you noticed getting a job has been getting more difficult? Do you have a different plan to get around this issue?