If you can measure a jobs productivity, you can probably replace that job with a machine, so when it comes to humans in the workplace we should be measuring different things. Photograph: John Macdougall\/AFP\/Getty Images<\/p>\n
A new report by the US-based National Academies of Science Engineering and Medicine suggests that not only has the automation of work barely begun but that the ways in which we measure the effects of technology on employment are inadequate to the task. <\/p>\n
The authors argue that to understand how automation is transforming our workplaces, we need better ways of tracking technological change. Put simply, they are saying that if we are what we measure that is, if policy is driven by the information we collect then we are collecting the wrong information. <\/p>\n
Data on many of these trends are elusive, reflecting [the] changing nature of society and the economy, and gaps in [the] statistical infrastructure, the report says. <\/p>\n
It points out, for instance, that we dont have a regular source of information about workers in part-time and other sorts of casual employment. Nor do we have good information about investment in computer technology at either the level of the company or of any given occupation. <\/p>\n
Also lacking is long-term information about the way in which skills within particular jobs are changing, as well as data on how effective educational practices are in preparing people for work. Such information gaps undermine our ability to respond appropriately to technological change and its effects on employment. <\/p>\n
This is a huge wake-up call for governments and businesses around the world who are proving slow to engage with the changing nature of work and who have tended to hide behind the mantra of jobs and growth, as if that will take care of everything. It is a reminder to all of us that we are long way from understanding what the future of work really looks like.<\/p>\n
The authors call for three new indices to be developed, tools that can be used to plug holes in conventional measures such as GDP, productivity and the unemployment rate a technology progress index, an artificial intelligence progress index and an organisational change and technology diffusion index. <\/p>\n
They set out the parameters of each in some detail and, in so doing, open up a much-needed discussion about the data used to help form public policy. <\/p>\n
We tend to think of measures like GDP and productivity as eternal truths of economics and, indeed, they have proved their worth over time. Nonetheless, some of them are not only reasonably recent inventions, dating from around the second world war, but are designed to measure activity in an economy of mass manufacturing, a sector increasingly being displaced by the information economy as the primary source of global wealth. This means the measures themselves are also increasingly irrelevant. <\/p>\n
As the economics professor Richard Holden wrote: The IMF model suggests Australian unemployment falling to 5.2% in 2017 and to 5.1% in 2018. But that is a pre-2008 model of how the labour market and macroeconomy interrelate. Maybe thats still the right model but I wouldnt bet on it. <\/p>\n
As the entrepreneur and founder of Wired Magazine Kevin Kelly has said on the subject of productivity: Productivity is for robots. Humans excel at wasting time, experimenting, playing, creating and exploring. None of these fare well under the scrutiny of productivity. That is why science and art are so hard to fund. But they are also the foundation of long-term growth. <\/p>\n
To help understand the point Kelly is making, consider that a quarter of Britains top actors have been kept in work over the last decade by Harry Potter films. So although JK Rowling may be a billion-dollar industry, her value as a contributor to national wealth does not improve by subjecting her to a stopwatch and increased output to improve her productivity. <\/p>\n
What Kelly is saying is that, if you can measure a jobs productivity, you can probably replace that job with a machine, so that when it comes to humans in the workplace we should be measuring different things. [Our] notions of jobs, of work, of the economy dont include a lot of space for experimenting, playing, creating and exploring, Kelly says, but those are the very skills that are likely to become more valuable in the workplaces of the future. <\/p>\n
So the value that humans will increasingly bring to the workplace is to be not a robot, which will mean measuring our contribution by something other than inputs and outputs. <\/p>\n
The National Academies report is not arguing for a wholesale replacement of traditional measures of economic activity but it is saying we need vast new supplementary data to better understand the ways in which new technologies are affecting the work that we do. Until we develop and implement these measures, it will mean that, on everything from education to welfare to employment policy, governments are flying blind. <\/p>\n
The concerns of the reports authors are being driven by their belief that the technological disruption of employment has barely begun. They write: Opportunities for digitising and automating tasks are far from exhausted. In particular, the workforce will be increasingly affected as more and more cognitive tasks become fully or partly automatable ... and as advances in robotics yield enhanced physical dexterity, mobility and sensory perception in machines. These trends will almost surely change the demand for the workers performing these tasks and the nature of the organisations in which they work. <\/p>\n
And so the sooner we start accurately measuring what is happening, the better. <\/p>\n
<\/p>\n
Read this article: <\/p>\n
We need to track more than GDP to understand how automation is ... - The Guardian<\/a><\/p>\n","protected":false},"excerpt":{"rendered":" If you can measure a jobs productivity, you can probably replace that job with a machine, so when it comes to humans in the workplace we should be measuring different things. Photograph: John Macdougall\/AFP\/Getty Images A new report by the US-based National Academies of Science Engineering and Medicine suggests that not only has the automation of work barely begun but that the ways in which we measure the effects of technology on employment are inadequate to the task. Continue reading