UBI (given strong identity) is actually not problematic in this way, assuming it's distributed fairly to everyone (and not just those who vote for it). If given to all unique participants in a flat manner, it actually acts as an anti-inflationary property, whereby it asymptotically flattens the wealth distribution. If done right, it's actually not a terrible solution to centralization of wealth -- but of course with other drawbacks.
For secret ballots, believe me, if it were trivially easy it would be done already. It's actually a pretty fun problem to try and think about how to do it and how to break it.
You can't just use a hash. Suppose Alice wants to vote for her 100 tokens for Bob. How do we prove 100 tokens voted for Bob without proving Alice was the one who did it -- while also ensuring Alice doesn't vote twice, or that she didn't move those tokens to a different account and vote there with the same tokens? How do we ensure Alice can't prove to Bob that she voted for him? How do we do all this in a deterministic way such that the whole public can verify the results and yet no one can prove who voted for who?
RE: How Vote Incentivization Monopolizes Delegated Proof of Stake