Read an interesting article in the penultimate Review of Higher Education that links "absorptive capacity" to scholarly productivity. Absorptive capacity was not a concept I was familiar with, but it refers to the ability to absorb new ideas. It depends on a lot of things, one being subject knowledge (which, one hopes, all faculty have). Scholars have different levels of "potential absorptive capacity," which they defined as ideas for research. This comes from exposure to others' ideas, as in when I go to an academic conference and come away with at least one research idea generated by reflecting on someone else's research, as well as personal factors (creativity, hard work). Then, various factors affect what they called "realized absorptive capacity" - i.e., productivity, but a big factor is organizational support. No surprise there, either; research funding and teaching load affect how many of those ideas one can make use of.
I should note here that this article only sets up a model; it does not test it. It sets up a series of predictions, many of which are unsurprising, i.e., institutional policies affect output.
The problem I see is that, in my experience, creativity has nothing to do with research output.
Yup, nothing. Scholars who publish a lot are very smart, very hard-working, and usually have institutional and personal support - but their approach to research is iterative and formulaic. In other words, they have standard operating procedures; the negative connotations of "formulaic" are unintended here. They have one standard methodology and use it to "mine" an area until that approach is exhausted in that area, then move on.
Real-life examples: A scholar who finds datasets and applies a particular econometric technique to variables that have been ignored in the past. A scholar who has a system for being the synthesizer of findings; once he/she has written the book on it, well, that's the book. A scholar who has honed a particular survey and analysis methodology and applies it to a series of related questions.
To put it another way - I'm allowed to use management-speak now, right? - they are hedgehogs that know one thing well, not foxes. Scholars who move across methods or topics (even closely related ones) don't publish as much. A hedgehog's first idea may have been highly creative, certainly, but the subsequent applications aren't. Once we've come up with the idea of a blind taste-test to determine consumer preferences for brands of strawberry jam, it's not creative to do the same thing with raspberry jam. It's not creative even if you move on to grape jelly after hearing the world's top tastebud scholar talk about grape jelly.
Simply, not all idea generation is creative.
Quick, make a list of things one can do on a date. Creativity will play a big role in how many different ideas you can come up with. But I'm betting that the number of dates you go on (or - even second dates) has little to do with how creative they are. As long as you can think of, "OK, coffee, dinner and a movie, and, um, dinner and drinks," you're doing fine.
Ergo: Take two scholars at the same institution and department, rendering the institutional variables irrelevant. Both have supportive spouses and no children. They have the same IQ and graduated from the same PhD program; both work hard. All that differs is their ability to come up with new research questions. I argue the more creative one will not publish any more than his or her colleague - because idea generation occurs much, much faster than follow-up. The minimum level of creativity necessary is met by practically every professor; more than that is irrelevant …
… except, perhaps, that too much might hurt your productivity. Spend a lot of time exploring new ideas or questioning things, and you're butting into the time required to research efficiently publishable ideas. A little creativity goes a long way: Too much, and you end up being Thorstein Veblen or China Mieville.*
Overall, I thought the model was interesting and probably useful; I just don't think creativity belongs in it.
* Given the choice, I would recommend "successful novelist" over "unable to hold an academic job," but your mileage may vary.