Perhaps one needs to distinguish between “stickiness” that derives from the client and “stickiness” that derives from outside counsel. If outside counsel is responsible, it’s probably not due to any rigidity in the drafting process—the notion of rigidity suggests that even if the drafter wanted to implement change, there’d be barriers to doing so. Instead, outside counsel tend to stick with a given form of contract because it’s the easiest thing to do, particularly when deal-volume is high and the bulk of the drafting is done by junior associates.

Economists have understood this concept for years; it's called "menu costs" -- the notion that a restauranteur may know she can charge higher prices, but won't if the cost of reprinting the menus exceeds the added revenue from the higher prices.

A contrived example, but the premise is omnipresent: change is not costless, and is only optimal if the marginal revenue from change exceeds the marginal cost (i.e., a sort of "Hand test").

Another ubiquitous example: optimal lease length for real estate in the face of uncertainty.

Another reason is negotiation costs. Many companies will leave contract terms vague or vaguely suboptimal because if they change it then the party on the other side might explicitly require removal, whereas vague claims allow future wiggle room.

I explore something similar in an upcoming article on patent claim drafting.

For empirical evidence on stickiness, I'd also recommend looking at these:

Kyle J. Mayer and Nicholas S. Argyres, "Learning to Contract: Evidence from the Personal Computer Industry" (Organization Science, 2004).
http://orgsci.highwire.org/cgi/content/abstract/15/4/394

Nicholas S. Argyres, Janet Bercovitz, Kyle J. Mayer, "Complementarity and Evolution of Contractual Provisions: An Empirical Study of IT Services Contracts" (Organization Science, 2007).
http://orgsci.journal.informs.org/cgi/content/abstract/18/1/3

Thanks all around for the good comments. Kip, I had not seen the term "menu costs," but that is a memorable metaphor. Peter, thanks for the links. I already had the first paper, but I hadn't seen the second.

I suspect much of this is unintentional and stems from the failure of lawyers to stay up to date with the law. I have seen this firsthand where corporate partners at major law firms don't stay on top of Delaware decisions (which is no doubt difficult when you have to tend to paying clients) and don't realize that their forms need to be updated (same thing happens frequently for bylaws and charters). Still speaking anecdotally, I believe this is less prevalent in niche areas like venture capital (most VC agreements that I see now account for Benchmark, Avatex, etc.) and more likely to occur with corporate "generalists" (e.g., in stock purchase agreements).

As a lawyer principally doing complicated contract-drafting (and a former student of professor G.'s contracts section), here's my take:

1) Clients frequently don't want new contracts. They want one in which they can fill in the business terms and run. This is an attempt to lower transaction costs all around (if for no other reason, it means avoiding retraining the business people and foregoing a lawyer at the early stages). Therefore, those contracts tend to be of two flavors depending on the other side: lopsided with highly favorable terms (think things like EULAs or other client strong negotiating position) or suboptimal and unlikely to draw much objection/negotiation (level playing field business relationships). The latter is probably what this article describes.

2) Clients don't like how expensive it is to have a lawyer be like a secretary. Whether there's actual truth to that idea is debatable, but that's certainly how some feel. Nevertheless, making every provision optimal is expensive. Even making most provisions optimal is expensive. Things get more expensive if it forces negotiation.

3) Most clients don't ever exercise the optimal contract provisions unless things go really badly. Now your example, "exclusive" v. "non-exclusive" seems like bad business/lawyering. However a number of suboptimal provisions remain because clients simply don't care and they see the 1 in 100 chance of exercising most provisions as negligible cost of doing business anyway.

4) Lawyers don't like to draft contracts.

In The Sun Printing and Publishing Ass'n v. Remington Paper and Power Co., Cardozo refused to enforce an agreement for the sale of paper because the contract was missing key terms. I have a very precise (but perhaps inaccurate) recollection of my contracts professor telling us that the contract at issue was a standard form contract used in the paper industry at the time. Apparently some legal scholars had looked into the matter and discovered that, despite Cardozo's ruling, the industry persisted in relying on the same form contract in subsequent transactions. Any further details escape me now, but I should be very glad to be reminded of them if anyone has more ready familiarity with the case and with the scholarship at issue.