There's not one issuer. In the US there are 51 issuers of car titles alone (50 states + DC) not counting the territories. Plus while the title is signed but not registered with the new owner it's a bearer instrument. People wash titles by taking them out of state because there's no effective way of binding both title histories together.
Again though what benefits are offered by adding the complexity of a blockchain compared to simply having a shared central database? Either way to be effective you have to get all the parties involved to agree on using one system, that's the actually hard part, the technical side is trivial by comparison.
The only situations where blockchains really make sense are when you have multiple parties involved who don't trust each other AND are unwilling to agree to trust a third party. Digital currency is basically it. Almost anything else is better served by a traditional database or append-only log. Certificate Transparency provides a good example of a more efficient way to do things when you're willing to at least partially trust certain authorities.
That's the right argument for the wrong customer. The end user is supposed to trust the government to act as one - that's the point. But a government is always in a state of partial synchronization with respect to record-keeping. That's why there are delays and loopholes in the system. And the naive answer of "just trust the one database bro" immediately creates gridlock: for the number of services that government provides, all pragmatic answers instantly turn into Byzantine General negotiations: who hosts the servers, who gets access, what kind of access do they get, whomever has the most stringent uptime and disclosure requirements will push the others into adopting the same, and so on...
On the other hand, if you say "it's append-only and distributed at the top level and every agency can put what they want on top of that, add new nodes at will, etc.", you have a blockchain. It redefines the desynchronization problem in a way that narrows the scope of the technical/political crossover. Negotiations still take place, unnecessary duplication and desynchronization still happens, but basic issues around where the global source of truth "is" are no longer a factor.
I don't think the effort described is even a perfect match for the tech, but it's the kind of "wrong in the right direction" needed to move forward in studying the issue.
