This leads to the truly disturbing "dirty little secret" of the publishing industry: the so-called "house name." Perhaps the most famous of these is "Franklin W. Dixon" (The Hardy Boys), which was actually half a dozen or so writers; "Carolyn Keane" (Nancy Drew) isn't quite as diverse, but had no more existence. This becomes deceptive when one realizes that publishers' treatment of the author as a "brand name" is not the only meaning of "author." To say that this lends itself to abuse and misuse is a serious understatement. Once upon a time, this involved prejudices in publishing against gender (George Eliot), religion, ethnicity, and so on. Ironically, the biggest culprit now is probably category-romance fiction, as a substantial proportion—perhaps even majority—of the men who write those works do so under a feminine or indeterminate pseudonym.
The real fun starts when, purely for business reasons (that is, previous sales were disappointing to someone), an author is coerced into writing under a pseudonym. In that particular instance, the "brand name" chosen is deceptive, especially when it is not somehow disclosed to the potential purchaser. I know to check the copyright attribution; if you're serious about publishing, perhaps you knew to do so before making your way here; but the majority of the bookbuying public does not, let alone the occasional casual browser.
The key is a question of intent. If the purpose of changing the author's name is to deceive potential buyers as to the identity of an author who has already published comparable works, it seems to me that one must question whether that constitutes a deceptive trade practice. Or, more strongly, not just question…