4 ms·
It may be lavish, but if an executive paid $350k is 35% less likely to make a mistake that would cost the organisation $1m (which is about 1% of their budget),
by dane-pgp 4y ago
It may be lavish, but if an executive paid $350k is 35% less likely to make a mistake that would cost the organisation $1m (which is about 1% of their budget), compared to a volunteer working for free, then maybe, from the organisation's point of view, the expenditure is cheap.
Of course, judging performance like that is very difficult, and predicting it in advance is even harder, so it's possible that the highly paid executive would actually perform worse than a volunteer (or a random number generator), but if the complaint about "lavishness" is really about inequality (i.e. the executive's standard of living being much higher than they need / the median citizen's) then that criticism should probably be directed at the tax policies of the relevant governments.
- 0xbadcafebee 4y ago> from the organisation's point of view, the expenditure is cheap But from any normal person's perspective, it's expensive. The difference is who is in control and what are their priorities and influences. Since "the organization" is making the decisions - and, completely incidentally, "the CEO" is the head of "the organization" - it just so happens that "the organization" finds that "the CEO" should be paid lavishly. Rich people gonna prioritize rich people.
- yupper32 4y ago> But from any normal person's perspective, it's expensive. Why does it matter what that hypothetical "normal" person thinks? Does that "normal" person have insight into how much it costs to hire a competent executive?
- JohnFen 4y agoIt matters because "normal" people are the ones paying those extravagant salaries in the end.
- yupper32 4y agoThose normal people should know that $350k is a steal.
- 0xbadcafebee 4y agoThe normal person knows that just having big paycheck does not make an executive more competent. We have all seen people with huge paychecks fail spectacularly.
- themitigating 4y agoAre you saying there's no correlation between salary and performance?
- fluoridation 4y agoThere's no direct correlation, yes. If you took two executives (or really two people in any profession), one making 10x as much as the other, and swapped their positions, you would not see one position dropping to 10% performance and the other rising to 1000% performance. In all likelihood, they would both drop to 50-70% for a while and then stabilize to their original values. Individual persons are not orders of magnitude more productive on their own, they're just in environments that allow them to be more productive, for example by giving them control over more resources.
- zihotki 4y agoI don't see any reason for an executive to be less likely to make a mistake. And considering that the core business is rock solid and didn't change much in last many years, I don't even see a potential for such mistake.
- JohnFen 4y ago> if an executive paid $350k is 35% less likely to make a mistake that would cost the organisation $1m (which is about 1% of their budget), compared to a volunteer working for free, then maybe, from the organisation's point of view, the expenditure is cheap. True, but I'd be hard-pressed to believe that's a realistic hypothetical at all.
- akolbe 4y agoSome of the most expensive WMF execs have been complete disasters, exiting after a year or two and leaving the entire organisation in disarray.
- yupper32 4y agoMaybe they should spend more than $350k and find someone more competent.
- akolbe 4y agoThat reminds me of the saying, "We have the best politicians money can buy."
- themitigating 4y agoCan you provide an example?
- Dylan16807 4y ago> It may be lavish, but if an executive paid $350k is 35% less likely to make a mistake that would cost the organisation $1m (which is about 1% of their budget), compared to a volunteer working for free, then maybe, from the organisation's point of view, the expenditure is cheap. They're not though. Especially not multiple of them providing the same service.