Is this time really different? Will Automation kill off development?

December 21, 2016

     By Duncan Green     

Is this time really different? That’s the argument whenever people want to ignore the lessons of automation-headlinehistory (eg arguing that this particular financial bubble/commodity boom will never burst) and such claims usually merit a bucketload of scepticism. On the other hand (climate change, nuclear war) sometimes things really are different from everything that has gone before.

Which brings us to technology. Lots of musings are circulating about the rise of Artificial Intelligence, automation etc. Driverless cars will put millions of drivers out of work. Robots will kill off manufacturing jobs. Everything will change.

At the World Economic Forum, Klaus Schwab talks of ‘the fourth industrial revolution’. The bible is the Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies,  a 2014 book by Erik Brynjolfsson and Andrew McAfee. Even President Obama has caught the bug, in a recent profile in the New Yorker

‘At some point, when the problem is not just Uber but driverless Uber, when radiologists are losing their jobs to A.I., then we’re going to have to figure out how do we maintain a cohesive society and a cohesive democracy in which productivity and wealth generation are not automatically linked to how many hours you put in, where the links between production and distribution are broken.’

aiWhich all raises a whole series of questions – is it true? If so, is that a Good/Bad Thing and for whom? Much too substantial for a blog post, but here are a few thoughts and links.

Is it true? Ever since the Luddites, the machines have been about to take our jobs (see this 200 year timeline, c/o Ranil Dissayanake). Techno-optimists argue that the current concerns are just another loss of collective nerve – new jobs will emerge to fill the gaps created by technology. But beyond the lessons of history, I see no reason why this should inevitably be true, at least not with the wages and conditions people have become used to (or aspire to). The scale and pace of the changes described above do suggest this time may really be different – a lot of disruption and job displacement could take place, at a time when governments seem less willing/able to help people adjust and cope.

Is that a good thing or bad thing? If the jobs being lost are menial or demeaning, shouldn’t we join those who are celebrating? Couldn’t this be the start of a brave new world where everyone works for 12 hours a week and then enjoys their leisure? (After all, that was what we were being promised back in the 1970s – back in 1930 if you read Keynes – yet somehow it never arrived.) For automation to usher in an age of positive leisure, an awful lot of public action needs to take place – states need to educate their people, those people need to organize to ensure that the benefits of automation are fairly distributed.

Because the alternative is rising inequality, captured in the Obama quote. At the extreme, something akin to HG Wells ‘The Time Machine’ and numerous other scifi dystopias – a privileged techno elite ruling over a mass of surplus-to-requirement proles whose main role is as threat to elite power.

And what does all this have to do with development? A lot of the aid business doesn’t seem to have the_revolutionnoticed these debates. It bangs away about the need for Industrial Policy and decent formal sector (preferably manufacturing) jobs – an essentially Fordist/Post World War Two agenda.

But there are a few interesting discussions. The ODI has started a fascinating piece of work on the implications of the gig economy for domestic workers in developing countries. Dani Rodrik argues that this is part of a wider story of ‘premature de-Industrialization’, and worries that the historic engine of development is running out of steam. IIED’s Andy Norton argues that green growth could be the answer to Rodrik’s dilemma. UNRISD has been gamely plugging away at researching one potential alternative – the ‘social and solidarity economy’.

Automation’s separation of paid labour from human wellbeing is also one argument for a Universal Basic Income, especially if paid for by taxing global public bads like carbon emissions. The UBI becomes the starting point, and the rest of the economy is laid on top of that. Would that address poverty? Possibly in the narrow income definition but not if we extend the definition to power, access and other dimensions. It certainly wouldn’t address exclusion, or discrimination, violence,

Time Machine, with Keith Richards lookalike

Time Machine, with Keith Richards lookalike

insecurity etc. If the UBI is only about rights to cash but no responsibilities, then we have something to learn from those who have been raised in a handout economy. Indigenous Australians called the dole ‘sit down money’ and hence developed the Community Development Employment Programme (basically welfare payments in recognition of what people contribute to the community, outside the formal market economy – eg caring for country, preserving traditions, etc).

What worries me more generally is the lack of politics. Lots of big picture, visionary, ‘If I ruled the world’ solutions, but what are the political conditions for bringing them about? The underlying story of the struggles of European Social Democracy is the decline of organized labour – and automation will further weaken that historical foundation for progressive ideas and policies. Attempts to rebuild progressive coalitions based on a ‘rainbow’ of different marginalized groups have had limited success to date. The danger is that, without a strong and coherent political voice, the future will be more HG Wells than UBI-supported leisured masses frolicking in the meadows.

Wow, I definitely need that Christmas break, but wd be interested in your thoughts!

Back in January, hopefully rested, optimistic and more coherent (I can always dream, right?)

Thanks to Irene Guijt, Katherine Trebeck and Paul O’Brien for comments on earlier drafts

December 21, 2016
Duncan Green