Allocating Positions

It may seem weird for me to ask this now, after all these years, but how does your college allocate positions?

I've seen it done in a few ways, and I have my own preference, but I've never actually seen alternatives spelled out systematically.

The most common way that I've seen has been as a sort of spoils system for gladiatorial combat among administrators. Each dean curries favor with the VP over time, and when hiring time rolls around, positions are allocated in rough proportion to the political standing of the deans.

I'm not a fan of that system, for what should be obvious reasons. It's arbitrary, it has nothing to do with teaching or staffing needs, and it redirects employee energy away from the core business at hand and towards internal politicking. Yet it survives, probably because it satisfies the egos of some powerful people. Not for me, thanks.

The second most common way is historical. In relatively flush times, this means one-for-one replacements; in leaner times, it means a de facto "take a number" system. (I suppose this wouldn't work during expansionary times, but I've never seen expansionary times, so the point is pretty abstract. Maybe someday...) This doesn't seem like any kind of improvement to me, since at its base, it's still arbitrary. Who's to say that the historical allocation is still the best? Even if it made sense when it was established -- a big 'if,' to be sure -- circumstances change over time. Worse, it gives rise to a sense of entitlement at the department level, which adds an unnecessary layer of conflict when the needs of the college have changed over time. "They took away our position." No, the college put it where it was more needed. It was never 'yours.' But good luck making that distinction when it has been the de facto standard for years.

It's also possible to allocate positions 'strategically,' which in management speak usually means deciding what the next 'hot' area will be and pouring resources into it. The argument for this 'picking winners' approach is that a college can build up a given area pretty quickly, even without a lot of money. The downsides are several. It starves out other areas, it rolls the dice on a single judgment call, and in practice it usually dovetails with method 1.

I've heard, too, of colleges using something like quotas, and calling it 'evidence-based.' The idea would be to set numerical parameters -- a given ft/adjunct ratio, say -- and use new positions to get outlying areas into compliance with the parameters.

This method strikes me as less objectionable, since it reduces the relevance of the courtier and actually refers to conditions on the ground. But it's still a blunt instrument. On the faculty side, adjuncts are easier to find in some fields than others, particularly during the day. Ignoring that reality will lead to some very unfortunate results in short order. Some fields -- music, foreign languages -- will always have relatively high adjunct percentages, just because the subject matter is spread over so many subfields (instruments or languages). You wouldn't add up single sections of Japanese, Russian, and Arabic and combine them into one position. And some fields have stringent external accreditation requirements that essentially take the choice out of your hands. This method also falls apart when comparing unlike jobs. Given enough funding for one position, what's the basis for comparing a financial aid application processor, a reference librarian, and a biologist? There's no obvious common denominator.

My personal preference is a hybrid evidence-based/star-chamber method. Have a meeting in which the deans make arguments using the same criteria, then have a vote. Having to make arguments with reference to given criteria can smoke out the weaker claims, and a vote reduces the chances of any one person's arbitrary whims mattering. It's still flawed, though, to the extent that criteria are blunt instruments and that voting can fall prey to logrolling, wheeling/dealing, and the like.

Wise and worldly readers, I'm wondering if you've seen (or imagined) a more successful way to allocate positions. Any great ideas floating around out there?