I've responded to the GP about how the US study cited is misleading, but the graph here seems to be worse, as it seems to simply misrepresent the data in the cited source, "Goos, Manning, Salomons (2014; Table 2)". Looking in the paper that it appears to be referencing [0], Table 2 shows -- instead of the across the board decreases in share of jobs in the low wage sector with increases in the middle and high wage sectors shown on the graph -- increases in most countries in both the low (Finland and Luxembourg being exceptions) and high (all countries) sectors, and decreases in the middle wage sector (again, all countries.) The first sentence of the conclusion of the paper states: "The employment structure in Western Europe has been polarizing with rising employment shares for high-paid professionals and managers as well as low-paid personal service workers and falling employment shares of manufacturing and routine office workers."
I've responded to the GP about how the US study cited is misleading, but the graph here seems to be worse, as it seems to simply misrepresent the data in the cited source, "Goos, Manning, Salomons (2014; Table 2)". Looking in the paper that it appears to be referencing [0], Table 2 shows -- instead of the across the board decreases in share of jobs in the low wage sector with increases in the middle and high wage sectors shown on the graph -- increases in most countries in both the low (Finland and Luxembourg being exceptions) and high (all countries) sectors, and decreases in the middle wage sector (again, all countries.) The first sentence of the conclusion of the paper states: "The employment structure in Western Europe has been polarizing with rising employment shares for high-paid professionals and managers as well as low-paid personal service workers and falling employment shares of manufacturing and routine office workers."
[0] http://personal.lse.ac.uk/manning/work/ExplainingJobPolariza...