Not in absolute length, but relative to what they have to say. A few reasons:
(1) Many academic books are revised Ph.D. theses. When you write a Ph.D. thesis, one goal is to show your review panel that you have mastered the literature. The easy way to do so is to cite and discuss everything even tangentially related to your narrative. Even after editing for publication, many books retain lengthy discussions of other people’s work which could easily be reduced from pages to sentences, if retained at all.
(2) For books written by law professors, law reviews encourage (indeed demand) turgid literature reviews and overfootnoting, and attorneys often equate a “thorough” legal brief with a legal brief that addresses any possible argument that the judge may think of–I remember spending hours researching a sentence or two for a footnote on obscure issues that the partner wanted to address “just in case.” These habits are hard to break when you write a book.
(3) Many professors aren’t writing for their readers. They are writing to impress tenure committees, to join an intellectual conversation followed by only a handful of others, or to create the equivalent of a reference book (as opposed to a book that you expect someone to actually read cover to cover). I suppose if these are the goals and they are satisfied, the book isn’t be “too long” from the author’s perspective. Indeed, there is a risk that a short, concise, clear book won’t be seen by one’s colleagues as having sufficient gravitas.
In my own books, I try very hard to write clearly and succinctly, to eliminate or at least limit tangents that distract from the overall narrative, to use footnotes primarily for citation purposes and not to make side arguments, to condense material as much as possible, and to write for as broad an audience as is consistent with both potential interest in the topic at hand and a reasonable level of sophistication. In short, I try to write for my readers.
I’ve been very pleased that some reviewers have appreciated my efforts with regard to Rehabilitating Lochner. A new review by Joseph Tartakovsky in National Review (unfortunately behind a paywall, but available to those with access to Lexis-Nexis) states that “Bernstein writes in a plain, clear style, and moves his story along at a brisk pace. This is a slim volume (though the small type makes it appear slimmer than it actually is), yet he manages to course through a century of shifting, complicated case law.”
An anonymous Amazon reviewer–no, I have no idea who this is–writes that “This slim volume is pithy and thought provoking. . . . He traces the crippling of the doctrine of liberty of contract in chapter 6, and does a much better job in that single chapter than others have done in much longer works. He has single sentences that convey as much information as many law review articles. . . . It’s incredible you get this all in only 129 pages!”
Maybe one day I’ll write a post about Rehabilitating Lochner’s editing process. But just to give an example, I eventually boiled down what was originally many pages about Cass Sunstein’s influential understanding of Lochner into two paragraphs. I could have easily written a book twice as long, that wouldn’t have been half as “good” from the reader’s perspective.
The vast majority of academic books I’ve read should have been at least 30% shorter, though there are some very long books–like Michael Klarman’s From Jim Crow to Civil Rights–that fully justify their length.
UPDATE: I’ll always be grateful to Professor Black of Brandeis University’s History Department, who set a strict five-page limit on our papers for his class on the French Revolution. For my first paper, I managed to cover the history of Jews during the French Revolution in those five pages, editing down 20+ pages of draft material into a paper Black told me was well-received by the one of the leading historian of French Jews (unbeknownst to me, his wife!). I learned from this experience that often pages could be cut down to paragraphs, paragraphs to sentences, and sentences to words, without the reader losing any essential information.