But it is Mongoose et all who are deciding not to de-logo their products.
I think you're also assuming that it's as simple as them firing up the production file (say for example an InDesign file) spend a half hour or hour removing any references to d20 STL stuff, delete the d20 logo, and they're done.
It's not that easy.
If it was _just_ the logo it might be.
The first big assumption being made is that they've still _got_ the files that were used for the print run. I got the distinct impression (from the Grand OGC Wiki project) that this wasn't in fact the case. At least for a fair number of their older books.
I don't know about other publishers, but it wouldn't surprise me to find out that many of them have their print files either lost or incomplete in some fashion.
So there's a definite amount of investment in terms of time, which translates to money. And that's going into a product that's "dead" for a decent chunk of the hobby. Well, that chunk of the hobby that's even going to bother buying 3rd party books to begin with.
After you get over the hurdle of even finding the darn files, then you're going to have to make sure you don't have errata that contradicts or fixes the text. Because let's face it: if a company goes through and spends some time re-tooling their books and _doesn't_ incorporate fixes to known errors? People rip on them. "I don't see what's so difficult about including [blah blah] since they were already taking the time to... [whatever]"
So now you've got the hassle of finding the files, making sure you don't have to include fixes of some sort, and then removing all the offending stuff. And then making sure that you _include_ some sort of OGL version of stuff that's otherwise missing. Because if you don't, then people are going to rip on you for putting out an incomplete/unfinished thing.
And after all that? Chances are the layout is going to have shifted. Which means going back through page by page to make sure the text flow hasn't gotten messed up in any number of different ways.
Potentially all of that work for just _one_ product.
Oh and what about artwork? I've no idea what the contracts are regarding the text and artwork, but there's the distinct possibility that re-tooling a book would mean having to go through and re-purchase art. Something along the lines of them (Mongoose) having bought the art to use for that _version_ of the game and possibly subsequent reprints of the book, but not for a new edition; and the OGL version could be argued to be a new edition.
I suppose it's possible that the artists (and writers?) gave up any rights to the work, and Mongoose can reuse any and all of that stuff from now until Doomsday. But I think that's an awfully big assumption to make.
Yeah, doing something like that for a small press type person _might_ be worth it. But for a larger publisher, I can certainly understand why they figure it's just not worth their effort.
Especially when you factor in things like:
Rising costs of paper
Overall cost increases in publishing (things like having to do larger print runs because it's more expensive for the printers and so forth)
Rising costs of shipping
A continued struggle to get _any_ money from an rpg product, let alone a "dead" line
Sure, the companies that aren't retooling their existing product have some blame. They're not willing to invest the money. But it's not like already existing fans are going to rush out and buy the _new_ versions. Heck no. "Why should I buy the same book again?" is going to be the question. So the company has to decide if there's actually going to be enough _new_ people showing up looking for _that_ specific product to justify the expense of redoing it.
The kind of products where it _is_ a simple matter to retool the product? I wouldn't be surprised if they're product that 95% of people aren't buying in the first place.