If your economics correspondent reckons there is more money to be made by not being in your newspaper then you have to start thinking about your business model.
The point Dubner made in the blog post is, I think, a valid reason to leave. It used to be that Freakonomics was a book with a follow-up blog trying to build/retain an audience for book #2. In that context, getting NYT to provide the platform makes a lot of sense. But when you're tossing in a whole lot of multimedia stuff and films and podcasts, it starts to make sense to run your own show.
Sort of like when you're starting up, it might make sense to run stuff on VPSes and shared servers and cloud or whatever, but as you scale, the incentives to own your own boxes triumph.
It could be that the business model is less suited to correspondents who have their own brand. Bringing the this particular blog into the NYT fold probably helped both a little but I am betting that the ongoing costs to the Freakonomics brand were getting to be a bit much (e.g. myself and probably thousands of others who dropped the freakonomics blog when the stopped providing full RSS feeds.)
Freakonomics is not in the newspaper. The newspaper refers to the print edition. Freakonomics is however in a profitable and highly trafficked web site.
The NYT web site business model works pretty well, but not enough to support the entire newsroom operations. However, the print side of the company still creates revenue to cover all that. The question is what business model will support that as print declines.
To your larger point I also disagree. They discuss their other ventures that they own (book, radio etc) which isn't owned or related to the Times. Therefore I'm guessing it's difficult for them to cross promote efficiently and drive people to different places. Moving to freakonomics.com makes perfect sense.
Sort of like when you're starting up, it might make sense to run stuff on VPSes and shared servers and cloud or whatever, but as you scale, the incentives to own your own boxes triumph.