It's simple economics. WOTC knows they made a mistake with the initial GSL, proven later when they revised it. They know they lost considerable market share to Pathfinder and are aware that Paizo's support of 3rd party products contributed to this (as well as their general public presence in general). But in the intervening years, several other companies have allowed free licensing for their rules (13th Age, Fate Core, Savage Worlds, etc), which has fractured a community previous under the umbrella of Paizo, previously under the umbrella of WOTC. If WOTC were to release 5E under OGL and offer licensing freedom similar to Pathfinder or the old 3rd Edition, it would suck all the 3rd Party Companies under the WOTC banner again.
I've made this unpopular assertion in the past, but I really find the benefits of an OGL to be debatable for Wizards.
For smaller publishers, I think the benefits are pretty clear. A smaller RPG company has many limits on their ability to grow their game, from a lack of marketing to the inability to fund a full product spread. An open gaming license means they lose some creative control, but the gains through having a greater presence should almost always outweigh any losses. When I see the 13th Age OGL, when I see Eclipse Phase hack packs and the company seeding their own torrents... those are great moves that seem to truly benefit those RPGs and their companies.
It isn't so clear for the very large companies. Wizards has a marketing department, even without dipping into the larger Hasbro marketing department. Wizards invests heavily in marketing and organized play programs. There aren't too many people in North America who don't know what Dungeons & Dragons is. Amongst gamers, how to find a game and where to find products are common pieces of knowledge.
We can look at the 4E product line and see a very robust product. It really isn't lacking. It didn't need third party products to complete the offering. Just about anything that we might suggest adding to the 4E product line would be a product that is unlikely to see large market share. And yet, if a third party had been allowed to create from the beginning, it might have dipped into something Wizards wanted to do. We know 5E will at some point a book on undead. But if another company publishes it first, Wizards could lose sales when they publish their version. Wizards just doesn't need help with either marketing or product spread.
The biggest proponents of an OGL tend to be freelancers, and our industry is full of them. Sure, if you got started by writing for d20, it can be hard to argue against an OGL. And some have argued that an OGL is necessary to keep great talent. I don't agree. Wizards continues to have a great staff with great talent. They continue to have great relationships with freelancers, including freelancers that also work for Paizo, MCG, and other prominent RPGs.
There is also the d20 glut and crash to consider. There was a lot of poor quality and it isn't clear that this was just an issue with companies learning the ropes. We can look at Kickstarter to see a lot of publishers and individuals making the same mistakes in judging their ability to deliver on quality.
The biggest argument against the OGL is Paizo itself. The OGL prevented Wizards, for the first time, from just moving onto the next edition. Paizo could use the OGL to take away WotC's customer base, keeping them on an older edition - and without a single penny to WotC. The OGL has further enabled the OSR movement and various other competitive initiatives.
It is interesting to see that Monte Cook didn't just replicate the OGL with Numenera, instead creating a separation by which a successful third party must negotiate terms. I would guess that MCG understands that they too have sufficient marketing and have plenty of ideas for products (witness The Strange Kickstarter). They want to engage people's creativity and create a community, but they don't want to see third parties usurp their product line or undermine their financial well-being.
For all those reasons, I don't think Wizards will use the same OGL as before. I can see some reasons for an OGL, but they are very narrow:
First, Wizards could go back to the very original concept of the OGL, which was that Wizards would publish core books and let the rest of the industry publish everything else. I don't think that's the case, however. Wizards seems still to want to publish more than just core books (witness their return to publishing adventures, starting with the excellent
Murder in Baldur's Gate).
Second, Wizards could decide to use a revenue sharing model, somewhat like the Apple store. I think that could have worked well with 4E. A way for anyone to publish their own adventure or supplement in an official store where it could be rated could be a way to allow everyone to benefit. Ratings could ensure quality and the store model could share revenue.
Thirdly, Wizards could be viewing this as the final edition of D&D or at least one to exist for a very long time. If that is the case, it could reverse some of what I wrote above. Because the normal product line approach won't be used (we aren't looking at just 4-5 years), it may make sense to allow more ways for third party publishers to use the material and to tap their creativity in keeping the rules and settings and adventures interesting. I really don't see why D&D Next would last longer than any other recent edition, but that may very well be the goal.
My personal hope is that the OGL resembles the MCG version, with perhaps more transparent revenue-sharing. I don't think a "just publish core books" model is really desired (the original concept) and the actual OGL sure seemed to hurt WotC. That seems to call for a new model where Wizards could ensure it benefits from third party efforts and still has come control to prevent enabling the growth of a major competitor.