Not that I know of, just describing what I've seen others push it for. With supply chain, in general, it has some potential value (for auditing, in particular) but for large things like cars and homes we have (usually) government mediated records, and they're also usually public or semi-public (accessible with some effort, but not freely available on the public Internet).

Title insurance might be cheaper if your county clerk’s title register were cryptographically verifiable by outside parties? Though that just requires a Merkle tree, not a blockchain

There is potential use for storing the history of real estate ownership in a more robust way, as part of doing a title search as a prerequisite for a real estate transaction. But that won't get rid of the need for title insurance, which exists exactly to offset the issues and dispute around ownership and liens against real estate that put undo burden on real estate transactions. And it's probably folly to encode into law that a single "blockchain" of real estate transactions is final and authoritative without any kind of legal review, which is also why title insurance exists. It would make things more robust and perhaps faster to verify, but it doesn't get rid of the social (non-technical) issues around real estate ownership transfer, either historically or at the time of the transaction.

> Is real estate fraud common enough?

I've talked to someone who worked on a (non-NFT-based!) solution for this in developing countries -- it turns out that yes, the cost to bribe officials to falsify documents to steal property can be cheaper than the property's market price.

> Is real estate fraud common enough?

No?