I would argue that this is a false dichotomy, so arguments for and against it would be misleading.
The observation is basically an overstatement, and it is due to our current confused notion of Kuhn's three modes of science: pre-science, paradigmatic negotiation, and normal science. From his perspective, these are successive stages in a cycle, we ascend from each into the next and then occasionally fall back to the earlier one when times get tough. But in reality, I would argue that this is an overstatement: they are all taking place all the time, especially in the younger sciences.
Starting at the 'top', in 'normal science' the rules are stable. Anything that does not follow them must be worked out, logically dodged, or put aside for consideration when other results elucidate the approach.
Questions are the important thing, then, because great insight is in finding out where the rules might be hard to apply or where working out some part of the canvas of explanation in detail might allow them to be applied to an open question. If the person working there knew why he couldn't proceed, he could probably proceed. He just hasn't had the insight to diagnose some gap, and if you just ask the right question, he can answer it and move on.
But at the same time, there are positional questions where things can go either way, where each of a number of alternatives must be worked out. Most obviously, these are the basic paradigmatic questions, where we spend whole centuries assuming one answer, say continuity of substance, and then find out it really loses to another, modified atomism, which then loses to another, 'vibrational' occupation of space by empty particles. Obviously, the right answer there, or the right support for it, would save everybody a century, and that can hardly be considered unimportant.
At the same time, these fork-and-find-out questions don't only arise at paradigm crisis points. They are a part of day-to-day science that really does not get credit in the classical definition of normal science as puzzle solving. There is always value in pursuing the less-favored options inside your paradigm. Working out those details can give a reframing or alternative approach that is still within the paradigm, but broadens everyone's toolset. When you are going down those paths, answers matter. Closing tributaries or finding alternatives are done with answers, not questions.
At the same time, parts of various sciences always remain in the prescientific state. There are underlying assumptions that simply fail to make intuitive sense, or to accord with natural philosophical notions. These questions are longstanding, almost perennial, they ask themselves, so the insight is not in the asking.
There too, creative answers bring things like the duality of wave and particle, or the underlying mystery of quantum state into the realm of application. As long as 'only wave' and 'only particle' were fixed for under-analyzed philosophical reasons, the duality remained a spooky, almost religious concept. A lot is gained in creative answers like those of QFT, which attach properties in new and interesting ways, or investigations that prove the underlying philosophical alternatives consistent with one another by injecting alternative notions.
So, yes, overall normal science is normal science, and the best approach to normal science is the most productive answer to what to do next. But there are payoffs for doing the exact opposite in both of the lesser modes of inquiry.