Objection 1: In most of the instances I'm aware of (please comment if you know of counterexamples), salary transparency only happens in places where compensatio

Objection 1: In most of the instances I'm aware of (please comment if you know of counterexamples), salary transparency only happens in places where compensation is based on a few obvious parameters like years of experience and job title, with little room for individual variation. In cases where there's a large variation in productivity between employees with the same role, it's not going to be possible to pay them commensurately in a system with fully transparent salaries. So this seems to only really make sense in roles where individual contributions are too tied up with group performance to disentangle, or where they're very hard to measure. Objection 2: To the extent that perceived relative pay is the variable that actually matters for determining how happy someone is with their compensation (ask someone in NYC versus someone in Mumbai with the same cost-of-living-adjusted salary how rich they each feel), then releasing salary information can be welfare-reducing even if it does cause salaries to rise in general. This is one of the strongest arguments opposing income inequality: the existence of very rich people makes everyone worse off by making them _feel_ poorer. Related hypothetical question: If every worker were paid exactly proportional to their actual productivity, how much would the standard deviation of labor earnings increase? (http://econlog.econlib.org/archives/2015/04/meritocracy_ble.html http://econlog.econlib.org/archives/2015/04/meritocracy_ble....)