4 ms·
>An offer of $x/yr salary and $b/yr bonus is less valuable to me than $x+n/yr salary, even if n<b and the bonus is characterized as a sure thing. In my experie
by owenmarshall 12y ago
>An offer of $x/yr salary and $b/yr bonus is less valuable to me than $x+n/yr salary, even if n<b and the bonus is characterized as a sure thing.
In my experience taking the bonus always makes more financial sense.
"No!" you shout in horror, "you're leaving money on the table because salaries compound and bonuses don't!"
Ah, but there's the thing: 10b is surely < 10n, but 3b is almost always > 3n. So while you're getting +n a year, I'm now making $[1.10 ... 1.25]*x/yr because I worked for 2-3 years and then found a new job.
Employers switching to bonuses gives a strong incentive for employees to try to secure raises by quitting (or threatening to quit). This strikes me as ridiculously counterproductive, but hey, that's how it seems to work, so you might as well ride the wave!
- djur 12y agoAll of that assumes that the bonus is actually paid. There are two cases where this isn't an issue: when the bonus is literally guaranteed (which tends only to be true of signing bonuses), or when the bonus is contingent on objective criteria you control, such as completing a particular project, or selling a certain number of accounts. Otherwise, a bonus is based on some combination of company performance (which is probably out of your control) and management whim. If management expects to issue the bonus every year, there's no rational cause to offer a larger bonus rather than a salary increase.