4 ms·
It seems to me, from reading the charts, that it still follows power law distribution, just with different cities. In other words, the labor dynamic that create
by yuy910616 5y ago
It seems to me, from reading the charts, that it still follows power law distribution, just with different cities. In other words, the labor dynamic that creates super star cities are still very much in place, just with new cities. Is that a fair reading?
Also, tangentially, what statistics should I calculate if I want to measure how "concentrated" the distribution is? The first thing that comes to mind is Gini coef, does that apply here or is there something else? The question I'm trying to answer is if the jobs are more dispersed among different cities or concentrated - assuming the cities themselves don't matter.
- ghaff 5y agoWell, city size (or really metro size presumably) follows some sort of power law-ish distribution in general. So, if you look at tech employers who historically mostly hired people from certain regions on the West Coast and see where they're hiring today, it's not surprising that they'd tend to cluster around larger metros given that's where a lot of people are. (Or at least there are more people in X large metros than there are in X arbitrarily-chosen smaller towns and cities.) I'm guessing to study concentration more broadly you really want to study a broader distribution of companies. I'm not sure that looking primarily at CA outflows for future job positions is necessarily an unbiased look at tech job distribution more broadly (and really not just tech jobs but employment at large West Coast tech companies in general).