Automation Redundancy Risk Is Fake News – Here's Why
The current messaging around technology adoption seems to be driven by three fairly negative statements:
- accountants and finance are missing out on all new technology benefits
- they are all doing something wrong or badly
- what we are doing is going to be automated; redundancy is imminent!
Whilst parts of these messages are possibly true, life isn’t so black and white, life isn’t changing at break neck speed and there is always another way of seeing the problem (and solution) especially if you know the profession as I do (and not just the marketing keywords).
Who’s to blame?
Much of this negative hype comes from the highly cited University of Oxford paper “The future of employment: How susceptible are jobs to computerisation?” by Carl Frey and Michael Osbourne, an excellent paper that categorises occupations based on their susceptibility to the advance of machine learning and robotics, and applies this to 702 profiles to rank which occupations are susceptible.
The trouble is… like any research project (read: Data Analytics project) there is assumption and detail that needs to be understood, for the consumer of the information to truly understand it, and make an informed decision about the information. Lazy analysis, leads to poor decisions.
UPDATE (02/01/2020) : Frey and Osbourne have released a statement correcting many of the poor conclusions people made about there research and the good news is they now state successful accountants are far less susceptible to automation than before! Link here
The marketeers approach…
…aka why let detail get in the way of a good headline…
Marketing says, “Someone told me about this report that was on the BBC website, its about when the robots are going to take your job! I looked up accountants and it said they are 94% at risk of automation! This is great as I need accountants and finance people to buy my new product, and this will definitely scare them in to action!”
Well, actually, no.
The fact is we are still very busy, still doing the job, still employed (in fact, the UK has never been more employed at 76.1% – ONS April 2019) and with all this technology noise and business uncertainty, sometimes its easier to add a few bums on seats, than it is to invest in a capital project like finance or analytics transformation.
But isn’t this ignoring the inevitable?
Maybe, but let’s look at those details…
When you read the research paper, the detail (that you need) paints a richer picture. Firstly, the hypothesis is based on whether it is possible to automate, not whether it is desired, beneficial, a priority. The research does not predict when our number is up.
The research is simply stating which jobs have the automation “opportunity” (my words), which characteristics are bottle necks to computerisation and what the risk looks like across the entire job market. There is a loose timeline mentioned by the researchers of 10-20 years, but that is highly speculative and the analysis does not address this.
And this is far from an exact science, like any data analytics project, the model is developed and trained using test data, 70 manually appraised occupations risk profiles (10% sample) then use to assess the 632 other occupations. Naturally, this will include an element of error, even though I do not challenge the findings as a guide.
It's also worth highlighting that the research concludes 33% of occupations are in the low risk category including the majority of “Management, Business and Financial” occupations.
I would add that we as humans, facing extinction or not, and flexible, adaptable beings, and finance professionals are educated, analysts and have a baseline skill set that is in demand and as much in the low risk range as the high risk range… we will adapt to survive!
We will decide when to make ourselves redundant, not the robots…
So yes, some accounting and finance jobs can and will be automated, but when that will be is TBC, and I would argue the WHEN is when we (as leaders) decide to automate, which will be when we (as a profession and finance team) have an “automation susceptibility mitigation plan“.
A plan that moves us to new, valuable areas of finance and accounting (ie: partnering, advising, analysing, communicating) and the less susceptible areas of the job market (problem solving, producing visualisation, team building, persuading and negotiating)… which most great accountants are already doing and will be doing more of.
Now, let’s not get complacent above transformation, as many of us aren’t changing, adapting and doing ALL the above to remain relevant, but I am confident we are not as ‘at risk’ as the hypers want us to believe.
Footnote: What the research actually says…
- Computerisation is moving from routine rules based activities, to cognitive pattern recognition activities.
- Physical robots are becoming more sensory and dexterous and will replace manual work.
- 47% of total US employment is in the high risk category
- 33% are in the low risk category including the majority of “Management, Business and Financial” occupations in the low risk category
- There is a correlation between level of education and risk of computerisation, higher education, lower risk
- There is a correlation between wages and risk of computerisation, high wage, lower risk
- Therefore, low skill and low wage workers are most at risk
Read Automation redundancy risk is fake news – here’s why, Part 2, where we prove that their is a path for us through this automation risk, and highlight the specific skills that will keep us on that path (spoiler: you don’t have to learn code!)
