Interesting take, but I think the end game is different. If there’s only one MKR holder, it’s true they can dictate an arbitrary MKR price. But it isn’t meaningful (compare to PTK). What matters is the extractable value (the extent to which they can actually cash out), which (w/o gov. attacks) is precisely future cashflows from stability fees.
Note: they also wouldn’t be able to borrow against the arbitrary MKR market cap b/c any well-designed lending platform wouldn’t (at least shouldn’t) accept such illiquid collateral at the arbitrary value. The arbitrary value chosen should have no effect on the game.
As far as I can tell (although this could be a fun research question), the only way this doesn’t happen is if the final MKR holder has a strange/irrational utility function. But relying on this would mean it’s not really decentralized.
Additionally, if there’s value from decentralization itself, the cashflow value could be negatively related to the dispersion of MKR. In which case the assumption that the game converges to a single MKR holder is probably wrong.
On the other hand, if we’re including value extractable from governance attacks, these are possible well before a single MKR holder is left.