I've heard this opinion worded slightly different many, many times over the years. I just can't agree with it. You're going to spend thousands of hours infront of a computer, it makes sense to invest time in being "efficient".
You reach a point of diminishing returns with everything. You can only gain so much knowledge/intelligence/experience before every increase in that becomes extremely difficult. Trying to become "smarter" when you are already "smart" is much harder than getting easy efficiency wins.
There are people that take it to the extreme of course, where they spend all their time on extremely tiny efficiency wins instead of learning how to program, but that's the same problem in reverse.
It's a pretty old concept, the first time I've seen it given a proper name was "aggregation of marginal gains" [0].