I recently run onto a study by the Brookings Institutions as I am preparing to give a talk on the future of work. The study focuses on the US but there are several similarities that would make it applicable to other parts of the world.
One of the thing that pops up from the study is that automation is going to have a significant impact on current jobs but the impact is significantly different depending on the job (see the interactive table they are providing).
The study points out that jobs involving people over 64% are the less likely to be affected by the automation (less likely is relative, though: they expect 40% of them to be affected versus 50+% of the ones involving young people). In a way getting older has some positive sides (not many, I can tell you from experience): the jobs you are likely to be involved in twenty years time will be jobs less affected by automation (this might also be a self fulfilling statement, since a jobs affected by automation will no longer be there in twenty years time and you would not be doing it, qed).
The study is also pointing out diversity based on ethnicity with Asian workers job slightly less affected.
However, both regarding age and ethnicity the percentage differences are not so high as to be really relevant. The real point the study is making is that present jobs have a half life of 20 years (50% of current jobs are going to disappear in 20 years time), with production, maintenance, food chain, administrative and transportation having significantly shorter half life. At the same time the study points out that automation will create new types of jobs. Although it is still to be seen if the jobs created will be sufficient to cover for the jobs lost (actually we need more jobs in the future as 2 billion more people will be walking on the planet by 2050), it is a given that the new jobs will require a different set of skills with respect to the old ones.
More than that. Someone is starting to show concern on the capability of humans to acquire new skills, both because of their complexity and because of the rate of change (by the time you eventually get the skill it may no longer be needed). This second concern -the pace of change- is actually the main reason that was given by Foxconn in 2011 when they announced the hiring of 1 million robots (it turned out they ended up in less than a 100,000 as of the end of 2018): they wanted to replace human workforce with robots not because the latter were less expensive (quite the opposite…) but because they were more flexible, faster to learn how to manufacture new products. As products life time was shrinking (and it is still shrinking) human workers could not keep up the pace.
Continuous education and sharing knowledge in mixed teams humans-machines will likely be the tools for success at both company and Country level in the next decades.