> I believe the effect of public salaries, all else equal, would be to move the average a bit up, but also greatly lower the variance. It might, and that doesn

> I believe the effect of public salaries, all else equal, would be to move the average a bit up, but also greatly lower the variance. It might, and that doesn't seem to be problematic, to me. The arguments about individual negotiation and proven value and all that don't change when salaries are public. In fact, the "savvy individual negotiator" argument might even be more applicable in that case: if Alice's market value is really 2x Bob, Fob or Gob then she should be able to demonstrate that quantitatively and be able to use that data to negotiate her appropriate value. The argument about developers' ability : value relationship is also a bit specious. I'm not convinced there actually is a good way to measure that, and even if there were it's very likely to be context-specific (e.g. some developers are going to exceed at systems programming or low level library development but might be marginal at best when tasked with design and implementation of systems that rise outside of that scope).