Hi! Thanks for your reply. I assumed ascending-bid first-price auctions because they are currently used in MEV-Boost and are (implicitly) assumed in most research on slot auctions or Execution Auctions.
That being said, I would expect these results to hold for most auction formats since the key result “the second-highest expected value (Early auction revenue) is lower than the expectation of the second-highest realized value (Late auction expected revenue)” holds regardless of the auction format.
I would expect the results to also hold for Dutch auctions. Dutch auctions do not seem clearly implementable, though, since it is unclear to me how there would be consensus on the state of the auction at any point in time between slots.
Could you elaborate what your point is about the social welfare implications of Trusted Advantage?