I find myself stuck finding a position regarding the government’s
introduction of T-Levels1 and the Chancellor’s recent promise of
more maths education.2 There are a few things to acknowledge. The
first is that we need skilled workers and that young people should have the
opportunity to gain skills and pursue their ambitions. The second is, as a
lover of mathematics myself, I do not object to a more comprehensive push for
maths education.
The reason I find myself stuck is I must ask the question of whether
these initiatives are really building an effective future labour force? Goos
and Manning (2007) paper ‘Lovely and
Lousy Jobs,’3 does a great job of explaining this issue.
The long-held belief is that automation (which doesn’t necessarily mean
computers - automation has been happening for years, as we will come to see -
but often does mean computers) would first reduce the demand for low-skilled
work, as these were (apparently) simple jobs. The result of this automation
would be an up-skilling of the workforce, which would be wonderful. This is the
sentiment behind the government’s current strategy (which, I reiterate, I don’t
condemn), and they have good reason to think this way.
Automation is a very popular topic in business schools, and whenever I
am discussing it with my colleagues I raise the point of Adam Smith. Smith
spends a respectable portion of The
Wealth of Nations discussing the shift from an agricultural economy to an
industrial economy.4 I proclaim, in said conversations, that all
society need do is embrace change and implement the necessary skill-shift to
accommodate the new dynamic, in the same vein as Smith, and any automation
fears would be set aside (I am simplifying here). In this sense, the
government’s policy seems to come from a sensible place (Smith I mean, not me).
Yet Goos and Manning argue that rather than targeting low-skill work,
automotive processes target routine work, irrespective of skill requirement.
And if we think about it, this makes a lot of sense. Computers can do extremely
complicated and time-consuming mathematics almost immediately. In terms of raw
cognitive power, computers have humans beat every time. In terms of abstract
cognitive power, less so.
A good example of this is accountancy. Services such as Quick Books
threaten many of the small accountancy firms.5 This is because of
automation, yet many people would not classify accountancy as a low-skill
profession. However, it is one that can be automated with a certain level of
ease, and as such it has become susceptible to our digital overlords.
Whilst the consequence of low-skilled automation should be the up-skilling
of the workforce, the consequence of routine automation is that only those
roles that are physically and cognitively difficult for computers to perform will
see growth in human labour demand. This is the infamous polarisation of work
and the hollowing out of the middle class, where low-skilled but physically
demanding jobs such as a cleaning see growth, high-skilled and cognitively
challenging work such as a CEO remain in demand, and middling jobs fall away.6
Returning to the government’s policies, this is where my issue lies. The
Chancellor recently stated he would like the British people to have good jobs,
which is a noble but mathematically difficult ambition - we can’t all be CEOs.
Similarly, mathematical knowledge will always be useful, but it will not be a
distinguishing factor for many top jobs in a world where we all have impossibly
powerful calculators in our pockets. Technical skills perhaps offer a ray of
hope - for the time being trades such as electrician look pretty safe. But
again, there is finite demand for all these things.
We already see the problem of over-qualified and under-demanded people
in this country, given the number of graduates taking low-paid work.7
My fear is that a promise that would have worked previously is now just
offering false hope.
I say this as someone who has optimistically espoused the wisdom of Adam
Smith. The intention of the government is to train young people in the skills
needed for the digital economy, much like Smith advocated for the industrial
economy. Yet that wisdom needs to be dragged into the 21st century.
To answer this, I refer to The Second
Machine Age by Brynjolfssen and McAfee.8
In this book, the authors argue that computers and humans are greater
together than as the sum of their parts. Where computers can process data in a
way humans never could, humans can apply the data and derive complex social
links that a computer would be blind to. Behind every good algorithm is an
insightful observation.
I have always been a supporter of a social contract for the digital age.
Automation and human work is part of that contract. The solution in terms of
up-skilling and re-skilling is to be more radical, and to embrace what we do
well which computers do not. T-Levels have merit, but alone they do not tackle
the growing polarisation of work. As part of a long-term strategy, the solution
is complex, though it almost certainly involves investment in technology,
robust institutions to ease the transition in work and a willingness to embrace
new ideas about education and skills-training. Our role in work is changing,
and so must we.
References
3 Goos, M, Manning, A (2007) ‘Lovely and Lousy
Jobs’ The Review of Economics and
Statistics, 89(1), pp. 118-133
4 Smith, A (2012, 1776) ‘The Wealth of Nations’ Wordsworth
Editions Limited, St. Ives
5 https://www.accountingweb.com/technology/accounting-software/can-software-really-replace-accountants
6 Autor, DH, Levy, F, Murname, RJ (2003)
‘The Skill Content of Recent Technological Change: An Empirical Exploration’ The Quarterly Journal of Economics,
118(4), pp. 1279-1333
8 McAfee, A, Brynjolfssen, E (2014) ‘The Second Machine Age: Work, Progress and
Prosperity in a time of Brilliant Technologies’ Norton & Company Ltd.,
London
No comments:
Post a Comment