Why the robot revolution risks an economic 'death spiral' for Australia | Greg Jericho | Business
In promoting his 10-year tax plan, Malcolm Turnbull suggested people want governments to undertake “long term planning”. However, a new research paper out this week from the IMF highlights how economies could be set for a major shake-up in the future and how sticking with the belief that better wages for workers comes from reducing company tax in order to spur capital investment is a rather wishful proposition.
Economics research papers generally are not known for their optimism, but the IMF paper titled “Should We Fear the Robot Revolution? (The Correct Answer is Yes)” fairly hits you between the eyes with its pessimism.
This research goes very much to the heart of primary political debate in this country about jobs, equality and the role of government.
Sign up to receive the latest Australian opinion pieces every weekday
When Malcolm Turnbull took over the prime ministership he loved to talk about how it was the most exciting time to be alive – innovation was on the rise, agility was all the go! But for many workers it is a rather worrying time. In the past decade since the global financial crisis real wages have stalled, underemployment has risen, wage growth has plummeted, businesses have sought to move away from enterprise bargaining, and the growth areas have been mostly in lower-paid services sectors such as social care.
And underneath all that is the feeling – perhaps overstated – that we are soon in danger of being replaced by robots.
The IMF paper on the topic notes that there are essentially two camps on this issue – the first is optimistic and believes that, as in the past, greater automation will see some jobs lost, but the demand for many jobs – especially those “that place a premium on creativity, flexibility, and abstract reasoning” – will grow and overall the economy is better off.
The other side of the coin are those who note these are not your grandparents’ robots we’re talking about. These are robots that make use of AI in order to do work previously believed to be non-automatable precisely because it was seen as creative, flexible, or needing abstract reasoning.
The paper considered a range of scenarios – from the more traditional one where robots replace only low skilled work to where robots are able to replicate a range of work and then a final one where robots “can do anything”.
And the results are not good for workers.
Essentially the shift sees national income move from labour to capital – as the returns from investing in robots to do work previously done by people increase.
They note that “the most common arguments for technology optimism do not stand up to scrutiny”.
Even in scenarios that fit with the more optimistic view of automation, the paper concludes that “automation is very good for growth and very bad for equality”.
The authors suggest that “in scenarios where the traditional technology disappears and robots take over the automatable sector, the economy either ascends to a virtuous circle of ongoing endogenous growth or descends into a death spiral of perpetual contraction. Unfortunately, the odds strongly favor the death spiral”.
It’s never good to read a report with “economic growth” and “death spiral” being used together.
And one reason to fear the future is the past.
Another report from the IMF this week looked at why some places in the US have had improved labour participation while elsewhere it has plummeted. It found that once you accounted for age, the overwhelming reason was automation:
Automation and offshoring may have permanently displaced some workers, even if their effects on the economy as a whole were beneficial, through the creation of job opportunities in other sectors or productivity gains”.It’s that rather awful nexus – things that improve overall economic growth and productivity invariably come at the expense of specific groups of people, and the gains which flow through to everyone can take a very long time to eventuate.
The authors looking at the benefits of increased robots concluded that the rise in automation causes real wages to fall in the short run but that they “eventually rise” due to increased demand for labour in work robots cannot do. But, they note bitterly, the “short run” can “consume an entire working life” and “eventually” could “easily take generations”.
It’s why basing your economic policy on “long-term” gains can easily fool you into pursuing something of little worth. And why in a period where we have seen strong employment growth, but flat household income, talk of cutting company taxes to increase capital investment is not a formula that is going to bring much joy.
Even the treasury’s own research suggests the company tax cuts will cause real wages only to rise in the “long term,” at least 10 years after they have been introduced – at which point we may well wonder what jobs will be around to garner that growth.
The most worrying part of the IMF paper is the authors have little sense that much can be done to ameliorate the problems. Yes, education can help, but even then the authors doubt a higher skilled labour force can offset the “huge real wage cuts unskilled labour suffers and the decrease in labour’s overall income”.
Similarly they note the proposal for a universal basic income has some merit but the problem is paying for it in a world where income is shifting from labour to capital. They note that “capital taxation is becoming more and more difficult with globalization and the increasing capital mobility”.
And it is also made more difficult when our government is following the lead of others to reduce the level of taxation that would be paid by those companies in the first place in order to encourage the investment that are causing so much worry for workers.
It has long been assumed such policies would work out better for everyone in the end; the research this week shows that that has not be true in the past and is not likely to be so in the future.
• Greg Jericho is a Guardian Australia columnist