The research on the relationship between intelligence and job performance
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.397.5384&rep=rep1&type=pdf
A few notes (some in this paper, some not):
[list]
[*]This is the general, albeit not unanimous consensus among psychometricians
[*]General intelligence correlates highly with every "type" of intelligence on the aggregate, though obviously there are individual exceptions
[*]General intelligence is a better predictor of job performance than years of experience or any of the Big 5 personality traits, though it should be noted that the latter's influence may be diluted due to personality tests being less precise than IQ tests
[*]Surprisingly, intelligence is a better predictor among candidates with *more* experience, probably because different starting experience levels even out over time.
[*]The more "complex" the job, the higher the correlation with intelligence.
[*]Some studies that suggest the opposite make the mistake in not adjusting for range restriction (e.g. employees within a role or educational bracket are likely to have a restricted range of aptitudes)
[*]Don't take this as an excuse to slack off because work ethic obviously matters, and might actually be more important (conscientiousness tests are less predictive than IQ tests but that may be because the test itself is less precise)
[/list]