I don't think this has anything to do with the IBM acquisition of Red Hat, but this is something that's driven by Red Hat's management itself. They did a similar thing with JBoss application server (the community edition) which was extremely popular and used within the community and even in production. Red Hat offered (and still offers) an enteprise version of it (JBoss EAP). Few years back they slowly started diminishing the prominence of the community version. They first renamed it (to WildFly) since the "JBoss" name confuses users with their enterprise edition offering. Then they started adding confusion around usage of the community edition of the server (now named WildFly) and started sending out messages (which never were clarified) that WildFly "cannot" be used in production environments (no explanation of what production environment was). Answers to these questions typically directed users to JBoss EAP (the enterprise edition) where they could enroll in developer programs and use the JBoss EAP for free for development use and then pay for the same when they use that or deploy that in production. This effectively killed the entire JBoss (now WildFly) community (external contributions, user discussions and any such interactions). WildFly these days is just there as a community project for the sake of it. There's no real external involvement in it and it's mostly driven by Red Hat employees and only rarely see any discussion happening in the open in their mailing lists anymore (can't blame them when no one external to their own employed team participates anymore).
The thing though is, developers actually involved in this project had strong opposition to the way this was handled, but none of it was heard and the "decision was already done" by people who hardly had any role to play on the day to day community involvement with the project. I am pretty sure it was the same with this CentOS decision.
There was even a restriction on the community edition to not release bug fix releases (just one was allowed). So if X.0.0 was released then X.0.1, X.0.2 and so on were not allowed. Without these bug fix releases the community versions started seeing X.1.0, X.2.0 and so on which included additional enhancements/features and weren't merely bug fix releases. As a result, the stability that the community edition of the server was known to provide (previously), no longer existed.
This all boiled down to one thing - the sales team couldn't convince customers that paying hundreds of thousands of dollars for the enterprise edition was a good thing when the community edition was equally good and well maintained, even if by volunteer community members.
The thing though is, developers actually involved in this project had strong opposition to the way this was handled, but none of it was heard and the "decision was already done" by people who hardly had any role to play on the day to day community involvement with the project. I am pretty sure it was the same with this CentOS decision.